--- Log opened Wed Feb 24 00:00:01 2016
--- Day changed Wed Feb 24 2016
20160224 00:00:01< vultraz> fabi: are you telling me the editor is three generations of work piled on top of each other D: 
20160224 00:00:13< fabi> at least
20160224 00:00:27< vultraz> oh dear god 
20160224 00:00:35< fabi> but I tried to get rid of some cruft
20160224 00:00:36-!- louis94 [~~louis94@91.178.241.57] has quit [Quit: Konversation terminated!]
20160224 00:00:47< fabi> all the hardcoded gui components for example
20160224 00:01:39< fabi> are gone 
20160224 00:01:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 248 seconds]
20160224 00:05:07< vultraz> well that's something
20160224 00:05:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 00:05:38< vultraz> well...
20160224 00:05:51< vultraz> EXIT_QUIT_TO_DESKTOP can be made to use qc::quit() in its current form
20160224 00:06:28< celticminstrel> A loop exited by means of setting a variable isn't particuarly odd.
20160224 00:06:48 * fabi thinks that also
20160224 00:06:56< celticminstrel> Gah, spelling.
20160224 00:07:13< celticminstrel> Honestly, it seems rather normal to me, when it's an event loop at least.
20160224 00:07:48< vultraz> well, I guess you're right 
20160224 00:08:24 * Aginor reads backlog
20160224 00:09:59< vultraz> play_slice() is a member of controller_base....
20160224 00:13:08 * vultraz groans
20160224 00:13:23< vultraz> the editor's exit status are handled by game launcher
20160224 00:13:59< vultraz> and in do_gameloop
20160224 00:15:14< vultraz> I think that case can be superseded by CVideo::quit
20160224 00:15:19-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 00:16:08< vultraz> but then I still need to do the suggestion from earlier 
20160224 00:16:54-!- kodomazer [62f86f38@gateway/web/freenode/ip.98.248.111.56] has quit [Ping timeout: 252 seconds]
20160224 00:16:56< celticminstrel> There could be a reason for the editor to get special handling with regards to exit status... not sure though, maybe not.
20160224 00:17:08< vultraz> I doubt it
20160224 00:17:17< celticminstrel> ...okay, do I have to kill XCode again?
20160224 00:18:25-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20160224 00:18:37-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20160224 00:18:37-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160224 00:19:52< vultraz> celticminstrel: in the case of "const boost::function<bool()>& prompt = &quit_confirmation::default_prompt", if I overloaded default_prompt this will still point to the non-argument one, right?
20160224 00:20:59< vultraz> don't know if i would, just seeing
20160224 00:21:58< celticminstrel> If you overload it (ie, same name, different arguments) then yes, it will certainly not point to your overload.
20160224 00:23:50< celticminstrel> But, perhaps you actually meant to say override. However, you can't override it, because it's not virtual, and it can't be virtual, because it's static.
20160224 00:24:42< celticminstrel> If you hide it (which is like overriding, but without virtual), then there's no relationship between it and your new function.
20160224 00:30:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 00:32:21-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 276 seconds]
20160224 00:38:29-!- gfgtdf [~chatzilla@f054060000.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 44.0.2/20160210153822]]
20160224 00:46:19-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160224 00:52:52< celticminstrel> What's with [modify_ai] appearing in side definitions?
20160224 00:53:07 * celticminstrel isn't quite sure whether it's directly under [side] or under [side][ai].
20160224 00:54:03< mattsc> celticminstrel: I am pretty much done with the page describing that:  https://wiki.wesnoth.org/Modifying_AI_Components
20160224 00:54:48< mattsc> From that page:  “Note that the [modify_ai] tag can be used both in [event] tags and in [side][ai] tags.”
20160224 00:55:16< celticminstrel> So. why would you want to use it in [side][ai]?
20160224 00:55:49< mattsc> I don’t know.  I assume somebody thought that would be convenient when they first introduced this.
20160224 00:56:06< celticminstrel> I see. Okay then, whatever.
20160224 00:56:14< mattsc> That’s how I look at it.  :)
20160224 00:56:15< celticminstrel> Maybe it would be convenient. I dunno.
20160224 00:56:37< mattsc> Well, it does make sense to have a way of modifying the default AI right at scenario start.
20160224 00:57:08< mattsc> This is how Crab (I presume) and whoever his mentor was chose to do that.
20160224 00:57:22< celticminstrel> Is there anything that can be done with [side][ai][modify_ai] that can't be done by simply specifying aspects and stuff in the [ai] tag?
20160224 00:57:53< mattsc> Yes, and no.
20160224 00:58:55< mattsc> You could do it all, but it is less code to do, say, “use the full AI and remove the villages CA” than “set up the full AI manually but with the villages CA missing"
20160224 00:59:07< mattsc> So my interpretation is that it was done for convenience.
20160224 00:59:12< mattsc> Without knowing that for sure.
20160224 00:59:17< celticminstrel> Ah, okay then.
20160224 01:00:11< celticminstrel> What's your opinion on the older [protect_unit] etc syntax for goals?
20160224 01:01:28< mattsc> If it significantly simplifies something in the code, throw it out.  If not, I don’t see why we shouldn’t keep it around.
20160224 01:02:35< celticminstrel> It's 14 lines of code to convert it to the standard [goal] syntax.
20160224 01:02:53< celticminstrel> I've assumed though that radius is specificed by protect_radius= instead of radius=, as was originally the case.
20160224 01:04:36< mattsc> ok
20160224 01:04:49< celticminstrel> ...DiD seems to use an undocumented [protect_area].
20160224 01:05:03< mattsc> If we ever go through that AI code cleanup we talked about, I would get rid of all this duplicate ways of doing things then though
20160224 01:05:13< mattsc> I see
20160224 01:05:39< celticminstrel> Scenario 7.
20160224 01:05:43< mattsc> There’s quite a bit of dysfunctional AI code all over the place, even in mainline
20160224 01:05:43< celticminstrel> Line 208.
20160224 01:06:11< mattsc> The problem is that most changes to the AI have quite subtle effects, so people often don’t know if they are actually working or not.
20160224 01:06:45< celticminstrel> If you prefer not having the secondary goal syntax, I can drop it. Unlike the simplified aspect syntax, it isn't really significantly different from the full syntax.
20160224 01:07:11< celticminstrel> I'm guessing the DiD one should be a goal with protect_location.
20160224 01:07:36 * celticminstrel suspects grouping=offensive also does nothing.
20160224 01:08:35< mattsc> grouping used to work the last time I checked, but that was a _long_ time ago
20160224 01:08:59< mattsc> I don’t mind keeping it around for now.
20160224 01:09:12< celticminstrel> I dunno, maybe it does, maybe it doesn't; but it wasn't in the list of aspects when I went through them last year.
20160224 01:09:51< celticminstrel> I've removed the version= key on my branch (didn't push yet, and haven't tested anything either).
20160224 01:10:21< mattsc> Umm, it (grouping) is there right now, I just checked.  In 1.12.
20160224 01:10:52< celticminstrel> I meant the list of aspects on the wiki. Sorry.
20160224 01:10:58< celticminstrel> You're right, it was listed in the code.
20160224 01:11:00< mattsc> The question is rather whether it is taken into account by the current MtT CA, or only by the one of the old default AI.
20160224 01:11:10< celticminstrel> I guess I missed that before.
20160224 01:11:23< mattsc> It’s also on the AiWML wiki page, and I am quite certain I did not change that in the last year.
20160224 01:11:40< mattsc> It would be possible to check the history for that, but I am too lazy for that  :P
20160224 01:11:43< celticminstrel> Oh, wait, it is in the list.
20160224 01:11:49 * celticminstrel sigh.
20160224 01:11:51< celticminstrel> Silly.
20160224 01:11:55< celticminstrel> Okay.
20160224 01:12:19< celticminstrel> So, the simplified aspect syntax means that any unrecognized tag or attribute within [ai] gets transformed into an aspect.
20160224 01:12:19< mattsc> As for the old target syntax, I’ve heard complaints from UMC authors (who also happen to be devs) that they use it so often in their add-ons that it would be a major PitA to change all that.
20160224 01:12:43< celticminstrel> Recognized tags are engine, stage, aspect, goal modify_ai, and the four old-style goal tags.
20160224 01:13:25< celticminstrel> Recognized attributes are turns, time_of_day, engine, and ai_algorithm.
20160224 01:13:25< mattsc> Probably — but of course the AI does not do anything with all of them, and only the valid ones show up in the config.
20160224 01:13:35< celticminstrel> I mean in my branch.
20160224 01:13:52< celticminstrel> The invalid ones would probably show up now, but the AI would do nothing with them.
20160224 01:14:17< mattsc> Oh, I see.  Hmm, that’s probably ok.
20160224 01:14:19< celticminstrel> Can a Lua CA check the value of an aspect?
20160224 01:16:09< celticminstrel> If so, this change would mean that Lua CAs can use custom aspects of their own.
20160224 01:16:15< mattsc> https://wiki.wesnoth.org/LuaAI#Functions_Returning_Aspect_Values
20160224 01:16:30< celticminstrel> Whee...
20160224 01:16:38< mattsc> Kind of fun, as I am updating all those pages at the moment  :)
20160224 01:16:43< mattsc> That one is not yet done though
20160224 01:16:49< celticminstrel> So this is a "yes, but not really".
20160224 01:16:55< mattsc> right
20160224 01:16:59< celticminstrel> Unless of course...
20160224 01:17:06< celticminstrel> Where are these Lua functions defined?
20160224 01:17:19< mattsc> in src/ai/lua, I believe
20160224 01:17:25< celticminstrel> core?
20160224 01:17:44< celticminstrel> Hmm, yeah.
20160224 01:17:44< mattsc> don’t remember; I just do a grep, usually
20160224 01:18:04< mattsc> Anyways, I’ll have to be afk for a while now.  I’ll stay online this time though.
20160224 01:18:13< celticminstrel> Looks like they're all hard-coded.
20160224 01:18:53< celticminstrel> As are the aspects themselves. Wow.
20160224 01:18:59< vultraz> celticminstrel: ok, I've got the stuff sorted
20160224 01:19:09< vultraz> do you think that loop is still needed?
20160224 01:19:09< celticminstrel> Though I guess it sorta makes sense, since each one needs to specify its type.
20160224 01:19:18< celticminstrel> vultraz: I have no idea.
20160224 01:19:49< mattsc> right
20160224 01:20:30< celticminstrel> So it needs more work before Lua can use custom aspects.
20160224 01:21:51< mattsc> celticminstrel: yes (and in fact, the current list of aspects in Lua is not even complete as far as the default aspects go).  However ...
20160224 01:22:06< mattsc> You don’t really need custom aspects in Lua.
20160224 01:22:24< vultraz> heh
20160224 01:22:30< mattsc> As in, you can easily have custom parameters, but they don’t formally have to be aspects.
20160224 01:22:31< vultraz> I actually broke editor loading
20160224 01:22:43< mattsc> The Micro AIs do that all over the place.
20160224 01:22:50< mattsc> anyways, now I’m really off for an hour or so
20160224 01:22:56< vultraz> fun fact: the editor hooks into the *same loop* that shows the credits at the end of a scenario 
20160224 01:22:58< vultraz> er
20160224 01:22:59< vultraz> campaig
20160224 01:23:01< vultraz> n
20160224 01:23:10< celticminstrel> Hmm, well, having aspects could make things convenient for other people using the AI, though.
20160224 01:23:30< celticminstrel> Since Lua is not C, there could easily be a general get_aspect function that returns an aspect of the proper type.
20160224 01:24:20< mattsc> Right - but what do you need an aspect for, that is not one of the default ones?
20160224 01:24:33< vultraz> do_gameloop() needs refactoring
20160224 01:24:42< mattsc> And those might as well remain hard-coded (since they are already, not because that’s elegant by any means)
20160224 01:25:00< celticminstrel> Well, you did say there were some missing though.
20160224 01:25:24< celticminstrel> So adding support for the missing ones could be done with get_aspect rather than individual functions.
20160224 01:25:26< mattsc> Yes, but those are not custom aspects
20160224 01:25:33< celticminstrel> True.
20160224 01:25:44< celticminstrel> ...provided the C code can fetch an aspect by ID...
20160224 01:25:48< celticminstrel> ^C++
20160224 01:27:41< celticminstrel> I don't know whether it's actually worthwhile to support custom aspects in Lua. It might be too much effort for too little gain.
20160224 01:27:56< celticminstrel> I think it would be worthwhile if it were easy to do, at least.
20160224 01:28:05< celticminstrel> But it looks like it's probably not easy to do.
20160224 01:30:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 01:30:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 01:34:16< irker437> wesnoth: Charles Dang wesnoth:master 7d0a02ec7d1a / src/ (7 files in 4 dirs): Bunch of refactoring of editor quit handling https://github.com/wesnoth/wesnoth/commit/7d0a02ec7d1af805e175f58d4adcbf3faef1b8ed
20160224 01:34:19< irker437> wesnoth: Charles Dang wesnoth:master 03ed7ffdff24 / src/editor/ (controller/editor_controller.cpp map/context_manager.cpp): Applied some formatting to Editor quit confirmation-with-maps-open message https://github.com/wesnoth/wesnoth/commit/03ed7ffdff24f5b9cfe7327b3b5b389c18cbb3dd
20160224 01:34:22< vultraz> celticminstrel: ^
20160224 01:34:29< vultraz> gfgtdf ^
20160224 01:37:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 01:37:48< irker437> wesnoth: Charles Dang wesnoth:master 031628dac5bf / src/editor/controller/editor_controller.cpp: Small Editor quit confirmation message wording change https://github.com/wesnoth/wesnoth/commit/031628dac5bfe95976e599735c2cd725e5ea095d
20160224 01:39:17< celticminstrel> mattsc: Regarding formatting, it might be good to be consistent with the other Lua pages? For example here, where I have the function in bold and parameters in italics. (There's no reason why my formatting has to be the final choice though - we could just as easily change everything else to match your AI page.)
20160224 01:39:18< celticminstrel> https://wiki.wesnoth.org/LuaWML:Units#wesnoth.get_units
20160224 01:41:04< celticminstrel> Does ai.synced_command do something different from wesnoth.synchronize_choice?
20160224 01:41:27< celticminstrel> Because, if not, I'd argue for deprecating it as passing Lua code as a string probably isn't the best way to handle things.
20160224 01:41:41-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has joined #wesnoth-dev
20160224 01:43:22-!- travis-ci [~travis-ci@ec2-54-146-170-187.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 01:43:22< travis-ci> wesnoth/wesnoth#8580 (master - f067332 : Celtic Minstrel): The build passed.
20160224 01:43:23< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111344386
20160224 01:43:23-!- travis-ci [~travis-ci@ec2-54-146-170-187.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 01:45:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 01:56:37< mattsc> celticminstrel: the formatting is still the original formatting Crab and/or nephro did from who knows how many years ago.  I have not gotten around to changing that yet.
20160224 01:56:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 01:57:03< mattsc> I agree that it makes sense to change it to what the other Lua pages use.
20160224 01:57:22< mattsc> There’s actually quite a few functions missing in that list and I don’t like the existing descriptions either.
20160224 01:57:28< mattsc> It’s next on the list.
20160224 01:57:43< mattsc> as for synchronize_choice, I don’t know right now.
20160224 01:58:37< mattsc> I need some way of executing arbitrary Lua code from a CA execution function
20160224 01:58:48< mattsc> in a synchronized way
20160224 01:59:05< mattsc> I think that’s what that function does, IIRC
20160224 01:59:31< mattsc> since ai.synced_command exists, I’ve never used wesnoth.synchronize_choice
20160224 02:01:52< ancestral> vultraz: It would be great it we had larger ToD graphics
20160224 02:02:43< mattsc> hi ancestral
20160224 02:02:50< ancestral> mattsc: Hey
20160224 02:03:09< mattsc> just a quick question, have you updated the OS X library packs yet for the SDL 2 switch?
20160224 02:03:20< ancestral> No, I need to
20160224 02:03:44< ancestral> I have to also update Boost apparently
20160224 02:03:51< mattsc> Okay, just wondering.  I’ve been gone for so long, I haven’t compiled in ages, I really need to get back up to speed.
20160224 02:04:21< ancestral> If you dl from SDL.org or whathaveyou and drop them in it works
20160224 02:04:25< mattsc> Boost, really?  That’s too bad, although I think it was not a big problem when I did it, it just took forever.
20160224 02:04:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 02:04:45< ancestral> mattsc: Or I can zip something together for you if you like
20160224 02:04:47< mattsc> Yeah, you said that.  I’ll do that when I get around to it.
20160224 02:05:06< ancestral> SDL2 is fan freakin’ tastic
20160224 02:05:09< mattsc> Nah, no worries.  I can do it myself, I was just not going to do so if you had already done it.
20160224 02:05:16< mattsc> Cool.
20160224 02:05:34< ancestral> Color cursor works nice
20160224 02:05:41< mattsc> nice
20160224 02:05:58< mattsc> It sounds like celticminstrel will want me to test some AI code soon, so I better get back up to speed.  ;)
20160224 02:07:36 * Aginor high-fives ancestral 
20160224 02:07:45< ancestral> Hey!
20160224 02:07:55< Aginor> hey back!
20160224 02:07:59 * ancestral found another bug
20160224 02:08:11< Aginor> I'm suddenly less enthused :D
20160224 02:08:32< ancestral> (Mini-map buttons)
20160224 02:08:38< Aginor> icons?
20160224 02:08:43< Aginor> on them
20160224 02:08:47< Aginor> they're wonky
20160224 02:08:55< Aginor> I think I've fixed it on the guifixes branch
20160224 02:09:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 250 seconds]
20160224 02:09:31< Aginor> that's an old bug as well, present in SDL1 builds
20160224 02:10:48< ancestral> Aginor: If you hide the app and come back, only two buttons show, and as incorrectly toggled
20160224 02:11:27< ancestral> This: https://img42.com/EQUfG
20160224 02:11:27< Aginor> ancestral: same behaviour on resize?
20160224 02:11:35< mattsc> Hi Aginor 
20160224 02:11:38< ancestral> Yes
20160224 02:11:49< ancestral> Same behavior on resize
20160224 02:11:50< Aginor> ancestral: yeah, that's an old bug :/
20160224 02:12:00< mattsc> I had a comment a couple hours ago following up on the ExpAI “discussion” we had yesterday.
20160224 02:12:00< Aginor> I might have inadvertedly fixed it though
20160224 02:12:05< ancestral> Cool
20160224 02:12:15< mattsc> Can’t stay around for follow-up right now though.
20160224 02:12:27< Aginor> mattsc: I don't have much time either
20160224 02:12:34< Aginor> I have leagues to run
20160224 02:12:40< mattsc> so long for now, then
20160224 02:12:46< mattsc> literally?
20160224 02:13:03< Aginor> mattsc: leagues and in sport leagues, not distance leagues
20160224 02:13:17< mattsc> oh, that’s disappointing  ;)
20160224 02:13:17< Aginor> as in even
20160224 02:13:25< Aginor> I'm working up to the other one
20160224 02:13:31< mattsc> cool
20160224 02:13:36< mattsc> well, ttyl for now
20160224 02:13:40< Aginor> extremely slowly
20160224 02:13:47< mattsc> that’s ok
20160224 02:14:04< Aginor> ancestral: I have inadvertedly fixed that bug
20160224 02:14:15< ancestral> Look forward to the fix :)
20160224 02:14:32< Aginor> I need to nail out a couple of more issues, then I can merge the branch
20160224 02:14:47< Aginor> as it stands, I'd fix a whole bunch of stuff and introduce some other regressions instead
20160224 02:18:51-!- un214 [~un214@2602:306:cccf:b439:56a0:50ff:fe57:101d] has joined #wesnoth-dev
20160224 02:21:15-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20160224 02:21:21-!- gfgtdf [~chatzilla@f054060000.adsl.alicedsl.de] has joined #wesnoth-dev
20160224 02:21:28< gfgtdf> compiling boost failed by lack of disk space ...
20160224 02:22:28< un214> what no ramfs /tmp?
20160224 02:22:37< un214> I can compile all wesnoth w/o writing to disk
20160224 02:29:33< ancestral> “Can” and “good idea” don’t necessarily go hand-in-hand
20160224 02:29:45< Aginor> that's actually not a bad idea
20160224 02:30:06< Aginor> if you can set your build-directory there
20160224 02:30:14< Aginor> but I don't know it yields much benefit
20160224 02:30:19< Aginor> it's not IO bound :D
20160224 02:30:31< un214> he ran out of disk space
20160224 02:30:54< un214> and if he has > 8GB ram he won't run out of RAM trying
20160224 02:31:11< Aginor> but if he wants to keep the output he's still got a problem
20160224 02:32:32< un214> possibly but it's the compiler intermediates that eat the most space
20160224 02:32:32 * Aginor shrugs
20160224 02:32:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 02:34:12< vultraz> ok, here's something I don't understand
20160224 02:34:30< vultraz> why is the titlescreen functionality specifically delegated to other functions outside its class
20160224 02:34:38< vultraz> besides twindow::get_retval_by_id
20160224 02:34:43< vultraz> there's also all the stuff in do_gameloop
20160224 02:35:11-!- travis-ci [~travis-ci@ec2-54-146-170-187.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 02:35:11< travis-ci> wesnoth/wesnoth#8582 (master - 27d4091 : Charles Dang): The build was broken.
20160224 02:35:12< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111353159
20160224 02:35:12-!- travis-ci [~travis-ci@ec2-54-146-170-187.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 02:35:56< vultraz> like
20160224 02:36:06< vultraz> why the hell does the main gameloop need to even encompass that
20160224 02:36:19< vultraz> just pass event callbacks to buttons in the dialog class 
20160224 02:36:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160224 02:40:22-!- travis-ci [~travis-ci@ec2-50-19-40-148.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 02:40:22< travis-ci> wesnoth/wesnoth#8583 (master - 03ed7ff : Charles Dang): The build was broken.
20160224 02:40:22< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111372014
20160224 02:40:22-!- travis-ci [~travis-ci@ec2-50-19-40-148.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 02:42:45< celticminstrel> vultraz: If it wasn't done this way, wouldn't that mean that the titlescreen dialogue is always open?
20160224 02:43:03< vultraz> I'm not sure
20160224 02:43:09< vultraz> but I'm hazarding a guess you might be right
20160224 02:43:22< celticminstrel> In any case, I don't think it's something worth changing.
20160224 02:43:32< gfgtdf> ok after removing 8gb of data i'll try again tomorrow
20160224 02:44:33-!- gfgtdf [~chatzilla@f054060000.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 44.0.2/20160210153822]]
20160224 02:46:07< vultraz> celticminstrel: well, there's a lot I'd like to simplify... move the titlescreen-related function to their own scope is just one... hotkeys shouldn't have tobe loaded manually by the titlescreen either
20160224 02:50:25< vultraz> sadly it's not a simple procedure 
20160224 02:51:49< vultraz> the gameloop is about as close you can get to the foundation besides main()
20160224 02:51:57< vultraz> which, incidentally, is what calls it :P
20160224 02:53:14< un214> did you see the program that started short main[] = { ... ?
20160224 02:54:30-!- travis-ci [~travis-ci@ec2-50-19-40-148.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 02:54:31< travis-ci> wesnoth/wesnoth#8584 (master - 031628d : Charles Dang): The build was broken.
20160224 02:54:31< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111372503
20160224 02:54:31-!- travis-ci [~travis-ci@ec2-50-19-40-148.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 02:56:34-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160224 02:58:25-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has joined #wesnoth-dev
20160224 03:07:30-!- un214 [~un214@2602:306:cccf:b439:56a0:50ff:fe57:101d] has quit [Remote host closed the connection]
20160224 03:19:04-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has joined #wesnoth-dev
20160224 03:20:23< mattsc> celticminstrel: Better? https://wiki.wesnoth.org/LuaAI#ai_Table_Functions_for_Executing_AI_Moves
20160224 03:20:58< mattsc> Not asking whether it’s good yet, it’s still incomplete and not fully proof-read, only whether it’s better.
20160224 03:23:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160224 03:26:23-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 03:27:05-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160224 03:31:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160224 03:42:15< mattsc> And, btw, I have not hit the limit on the DP song yet.  ;)
20160224 03:58:57< celticminstrel> ...a blacklisted CA?
20160224 03:59:13< celticminstrel> Yeah, it does look better.
20160224 04:01:34-!- prkc [~prkc@46.166.190.188] has quit [Remote host closed the connection]
20160224 04:01:42< mattsc> Yes.  That still needs to be explained elsewhere on the wiki.  And then I’ll link to there from that statement.
20160224 04:01:45< mattsc> Thanks.
20160224 04:01:49< mattsc> Still lots to do ...
20160224 04:01:59< celticminstrel> I feel like "status" could be better handled.
20160224 04:02:08< celticminstrel> On the API side, I mean.
20160224 04:02:14< mattsc> Feel free to change it.
20160224 04:02:41< mattsc> I doubt that anybody but me (of those who are left) even knew about this before I put it on that page …
20160224 04:15:24< celticminstrel> Well, for now, I'm just adding a "result" key which contains the string corresponding to the status code.
20160224 04:18:58< mattsc> Sounds great.
20160224 04:20:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 04:24:06< celticminstrel> vultraz: Do you know who can delete pages on the wiki?
20160224 04:25:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160224 04:26:22< mattsc> celticminstrel: vlutraz himself can do so.  And a couple others.
20160224 04:26:39< mattsc> s/vlutraz/vultraz 
20160224 04:26:51< mattsc> I do use autocompletion most of the time, but apparently not always.
20160224 04:27:12 * celticminstrel thought so, that's why I pinged him specifically.
20160224 04:27:23< mattsc> I have a whole list of old AI pages to delete, but I was going to wait with asking him until I am done with the restructuring.
20160224 04:28:05< celticminstrel> The FormulaAIandDynamicScripting page talks about "rulebases". I'm guessing this never actually got impemented?
20160224 04:28:21< celticminstrel> Or is it some obscure thing that was never properly documented?
20160224 04:28:30-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has quit [Quit: "The highest possible stage in moral culture is when we recognize that we ought to control our thoughts." - Charles Darwin]
20160224 04:29:28-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has joined #wesnoth-dev
20160224 04:29:49< mattsc> I think this was a GSoC proposal page from a student who probably did not get accepted.
20160224 04:30:34< mattsc> I think we should keep those for historical reasons.
20160224 04:30:48< celticminstrel> It sounds like other stuff on that page did happen.
20160224 04:31:02< celticminstrel> In fact, it sounds like that's where stuff like faifile...faiend came from.
20160224 04:31:18< celticminstrel> Though I didn't click the svn links to verify that.
20160224 04:35:26< celticminstrel> Can [micro_ai] appear inside an [ai] tag?
20160224 04:35:39< mattsc> okay; I have to admit I did not pay that much attention to that page yet
20160224 04:36:07 * celticminstrel is on AiWML now. Basically browsing through the AI category.
20160224 04:36:11< mattsc> I found it earlier today or yesterday when I was checking links to a page that I marked for deletion and added the AI category.
20160224 04:36:37< celticminstrel> I guess you haven't gotten to improving AiWML yet.
20160224 04:36:38< mattsc> celticminstrel: okay; as I said, not all pages are finished yet
20160224 04:36:52< celticminstrel> But the question about micro_ai?
20160224 04:37:07< mattsc> I have done quite a few improvements to AiWML, but it’s not 100% done yet.
20160224 04:37:24< mattsc> No, [micro_ai] is a standalone tag that goes into ActionWML
20160224 04:37:32< celticminstrel> Okay.
20160224 04:38:27-!- irker437 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160224 04:38:38< celticminstrel> The FormulaAI pages also double as reference for the formula engine used by GUI2 and filters.
20160224 04:38:55< celticminstrel> Though they're not very good at that, either.
20160224 04:38:58< mattsc> Yes — I have never liked that
20160224 04:39:04< mattsc> It is really not done very cleanly.
20160224 04:39:18< mattsc> But I am not going to clean those up, or I’ll never get done.
20160224 04:39:24< mattsc> So they’re all yours.  :D
20160224 04:39:27< celticminstrel> That's fair, it's not strictly AI-related.
20160224 04:39:46< celticminstrel> I wonder if FAI could be backronymmed to something that doesn't involve "artificial intelligence".
20160224 04:41:29< mattsc> I think it really should.
20160224 04:41:49< celticminstrel> Is ai_special= deprecated?
20160224 04:42:06< mattsc> no
20160224 04:42:23< mattsc> but it only has one value: guardian
20160224 04:42:27< mattsc> IIRC
20160224 04:42:30< celticminstrel> Apparently ai_special=guardian is the same as [status]guardian=yes
20160224 04:42:35< celticminstrel> For some reason.
20160224 04:42:54< celticminstrel> I would think ai_special is something that could be used by other AIs to attach arbitrary behaviour.
20160224 04:42:56< mattsc> It is; it’s actually kind of messed up, just as some of the other stutusseses
20160224 04:43:37< mattsc> when you set ai_special, it also changes the status; which means if you want to undo it, you need to clear both.
20160224 04:43:53< celticminstrel> Uh, okay then.
20160224 04:44:01< mattsc> There are a couple other things for which that is also the case, but I forgot what they are
20160224 04:57:00-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160224 04:57:52< mattsc> not AI-related things
20160224 05:02:16-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has joined #wesnoth-dev
20160224 05:02:51< celticminstrel> It sounds like the recruitment_instructions aspect does the same thing as the recruitment aspect, plus a bit more.
20160224 05:03:31< celticminstrel> This suggests that dropping the recruitment aspect and the recruitment stage wouldn't be a major problem, though I guess you'd have to get people to convert (since recruitment_instructions was poorly documented).
20160224 05:04:58< mattsc> The recruitment aspect and the stage are both from the old (pre 1.7) default AI.  I would get rid of them if (when?) we do the AI code cleanup.
20160224 05:06:02< mattsc> recruitment_instructions is clearly based on that, but it’s a newer and more powerful version.  That was done by flix during his GSoc project 3(?) years ago.
20160224 05:06:33< celticminstrel> The stage could be dropped first, with the aspect automatically converted to recruitment_instructions.
20160224 05:06:44< mattsc> sounds good to me
20160224 05:07:26< celticminstrel> As far as I know, the recruitment aspect only supports [limit], so converting that should be relatively easy.
20160224 05:07:56< mattsc> that’s consistent with my knowledge as well
20160224 05:14:37-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 05:19:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160224 05:21:34-!- irker475 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160224 05:21:35< irker475> wesnoth: sigurdfdragon wesnoth:master 7d2eaef38d4f / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): AOI: Make the named loyal units unnameable. https://github.com/wesnoth/wesnoth/commit/7d2eaef38d4fed1f10bdb2aad01ed2163f794270
20160224 05:21:35< irker475> wesnoth: Charles Dang wesnoth:master c9e2dd19db5b / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): Merge pull request #607 from sigurdfdragon/AOI_Named_Loyals https://github.com/wesnoth/wesnoth/commit/c9e2dd19db5b044a252708efbf3b6c6a97c1a3a1
20160224 05:21:42< vultraz> celticminstrel: do you need a page deleted?
20160224 05:22:00< celticminstrel> Practical guide to modifying AI, or something like that.
20160224 05:22:10< vultraz> link, please
20160224 05:22:17< celticminstrel> Bah. One moment.
20160224 05:22:30< celticminstrel> https://wiki.wesnoth.org/Practical_Guide_to_Modifying_AI_Behavior
20160224 05:22:38< celticminstrel> It also has a Chinese version: https://wiki.wesnoth.org/Practical_Guide_to_Modifying_AI_Behavior/ch
20160224 05:22:56< celticminstrel> I dunno whether you want to delete the Chinese version.
20160224 05:23:31< mattsc> vultraz, celticminstrel: please don’t delete any AI pages yet
20160224 05:24:00< vultraz> mattsc: ah, alright. was just about to
20160224 05:24:04< celticminstrel> Hmm, Create links to that page.
20160224 05:24:26< mattsc> well, I guess there’s a couple which have all content removed by now, but I want to make sure all links to them have been removed etc.
20160224 05:24:30< celticminstrel> mattsc: Sorry, since you marked it for deletion, I assumed you wanted it deleted.
20160224 05:24:37< mattsc> and I have not done that for all of them yet
20160224 05:24:47< mattsc> yes, I do, but see the above comment
20160224 05:24:49< vultraz> mattsc: link removal I can do
20160224 05:25:10< celticminstrel> mattsc: Should I link to Wesnoth AI instead from Create?
20160224 05:25:19< mattsc> vultraz: it’s not necessarily just link removal, some links need to be redirected to other pages
20160224 05:25:27< vultraz> mattsc: ah, ok
20160224 05:25:32< vultraz> i'll leave you to iy
20160224 05:25:34< vultraz> it
20160224 05:26:20< mattsc> celticminstrel: I’m in the process of doing all this
20160224 05:26:40< mattsc> I’m not saying I don’t appreciate the help, I do, but it’s actually easier if you just let me do it
20160224 05:26:58< celticminstrel> Fine, fine.
20160224 05:27:21< celticminstrel> That's the only important page that links to the now-missing Practical Guide, though.
20160224 05:27:28< mattsc> If you want to help, I am sure there are plenty o typos to be fixed etc.  I saw that you found some already.  Thanks!
20160224 05:27:34< irker475> wesnoth: Charles Dang wesnoth:master a87939acb02d / data/multiplayer/scenarios/ (2p_Dark_Forecast.cfg Random_Scenario_Lua_Cave.cfg): Run wmlindent https://github.com/wesnoth/wesnoth/commit/a87939acb02df8f6e3fa2ba626449a3574930963
20160224 05:27:38< celticminstrel> Yeah, I found some.
20160224 05:27:51< celticminstrel> And there's another somewhere... where was it...
20160224 05:28:13< mattsc> Yeah, I’ve already moved or removed lots of links to lots of pages, but I don’t have it consistently done for all yet.
20160224 05:29:02< mattsc> I actually have at least 7 pages marked for removal, but … well, it’s all still work in progres.
20160224 05:29:23< celticminstrel> Is "AI in 1.8" one of them?
20160224 05:29:29< mattsc> yes
20160224 05:29:50< mattsc> but there’s still some content left on it
20160224 05:30:24< mattsc> If you look at “Recent changes”, you can see that I’ve not exactly been idle  ;)
20160224 05:30:31< celticminstrel> Yeah, I looked there.
20160224 05:30:32< mattsc> I just only have that much time each day.
20160224 05:30:59< celticminstrel> AI Module lists the candidate actions using their symbolic name. RCA AI lists them using more user-friendly names.
20160224 05:31:01< mattsc> None of this is meant as a complaint.  Just give me a couple more days, please.
20160224 05:31:21< mattsc> Yep.
20160224 05:31:23< celticminstrel> Honestly, I don't like how the symbolic names work, but maybe it's too late to change that.
20160224 05:31:35< mattsc> Neither do I; and yes
20160224 05:31:54< irker475> wesnoth: Charles Dang wesnoth:master 1ff15d3ff168 / data/campaigns/ (22 files in 11 dirs): Ran wmlindent on campaigns https://github.com/wesnoth/wesnoth/commit/1ff15d3ff1680ca55e71741fde95b379fbf3eb25
20160224 05:32:11< mattsc> The RCA AI page is entirely new as of a few days ago.
20160224 05:32:19< celticminstrel> Oh, there's a page here about a hypothetical new stage.
20160224 05:32:45< mattsc> Well, not every single sentence, some were copied from other pages, but entirely new put together
20160224 05:33:40< mattsc> which one?
20160224 05:33:49< celticminstrel> Strategy Formulation
20160224 05:34:10< mattsc> Ah, yes.  That was an unsuccessful GSoC project.
20160224 05:36:51< vultraz> mattsc: ok, when you have a page ready for deletion, please file it under https://wiki.wesnoth.org/Category:Candidates_for_deletion and I'll get to it
20160224 05:37:19< mattsc> vultraz: sure, thanks
20160224 05:37:38< mattsc> celticminstrel: I am going to sign out shortly.  If I may, I’d suggest you give me a few more days.  That’ll also mean less work for you, because you won’t have to read something again because I changed it after you did the first (or second, or third) time.
20160224 05:37:49< celticminstrel> I was going to go to bed in a few minutes, too.
20160224 05:38:08< mattsc> good timing then  :)
20160224 05:38:49< mattsc> btw, I am not sure what to do with the Chinese version of two of those pages
20160224 05:40:17< vultraz> get rid of them
20160224 05:40:29< vultraz> if the english version goes the translations are useless
20160224 05:40:52< mattsc> well, the Chinese version is better than no version
20160224 05:41:16-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 05:41:18< travis-ci> wesnoth/wesnoth#8586 (master - c9e2dd1 : Charles Dang): The build is still failing.
20160224 05:41:18< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111398020
20160224 05:41:18-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 05:41:18< mattsc> I guess we could hope that somebody will eventually translate the new pages
20160224 05:41:35< vultraz> oh, what the hell is travis on about..
20160224 05:42:27< celticminstrel> tsynced_choice_wait::pre_show has an unused parameter.
20160224 05:43:01< vultraz> :|
20160224 05:43:26< vultraz> ah, yes
20160224 05:44:24< SigurdFD> vultraz: I think that'll be the last of my AOI PR's. thanks
20160224 05:44:49< vultraz> celticminstrel: btw, maybe my understanding of 'explicit' is wrong, but shoudn't the quit_confirmation constructor be explicit?
20160224 05:44:53< vultraz> SigurdFD: all good :)
20160224 05:45:11< celticminstrel> Uh, isn't the quit_confirmation constructor no-argument?
20160224 05:45:29< celticminstrel> Ohhh.
20160224 05:45:31< celticminstrel> Yes, it should be.
20160224 05:45:48< vultraz> ahh
20160224 05:45:49< vultraz> will do
20160224 05:45:55< celticminstrel> BTW, did you use a different prompt in the editor?
20160224 05:46:08-!- Kwandulin [~Miranda@p200300760F0BC5BCE849E64EDF47759E.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160224 05:46:27< vultraz> though this whole 'explicit' stuff seems weirdly arbitrary. Like, why did they sit down and say 'ya know, we'll make it so single-argument constructors need a special keyword'
20160224 05:46:30< vultraz> celticminstrel: yes
20160224 05:46:44< celticminstrel> Did you do that by passing it to the quit_confirmation constructor?
20160224 05:47:30< celticminstrel> Ah, yes, you did.
20160224 05:47:52< vultraz> gfgtdf did
20160224 05:47:54< vultraz> not I
20160224 05:47:56< celticminstrel> Ahh, okay.
20160224 05:49:19< irker475> wesnoth: Charles Dang wesnoth:master 5418ddfc2495 / src/gui/dialogs/ (synced_choice_wait.cpp synced_choice_wait.hpp): tsynced_choice_wait: removed now-unused CVideo parameter from pre_show https://github.com/wesnoth/wesnoth/commit/5418ddfc24955de2e83addb150b9dd6848bdfb71
20160224 05:49:57< celticminstrel> Uhh, you just broke everything.
20160224 05:50:05< celticminstrel> Probably.
20160224 05:50:27< celticminstrel> I forget, is pre_show pure virtual?
20160224 05:50:55< celticminstrel> Anyway, with that change, the pre_show will never be called.
20160224 05:51:21< vultraz> wait, really
20160224 05:51:26< vultraz> crap
20160224 05:51:55< vultraz> uh...
20160224 05:51:57< vultraz> ok
20160224 05:52:04< celticminstrel> Yes. Since the signature no longer exactly matches the one declared in tdialog, that function now hides the tdialog function instead of overriding it.
20160224 05:52:07< vultraz> what's the way to deal with unused parameters again 
20160224 05:52:28< vultraz> Remove the argument name from the header, or the implementation 
20160224 05:52:35< celticminstrel> Just remove the argument name from the implementation.
20160224 05:52:50< celticminstrel> You can remove it completely, or if you prefer, leave it in but commented out.
20160224 05:53:25-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160224 05:53:40< celticminstrel> (If we had the C++11 overrides keyword, that break would've been caught by the compiler.)
20160224 05:54:40-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 05:54:41< travis-ci> wesnoth/wesnoth#8587 (master - a87939a : Charles Dang): The build is still failing.
20160224 05:54:43< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111398712
20160224 05:54:43-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 05:54:54< irker475> wesnoth: Charles Dang wesnoth:master 096d257624e1 / src/gui/dialogs/ (synced_choice_wait.cpp synced_choice_wait.hpp): Redo 5418ddfc2495 properly https://github.com/wesnoth/wesnoth/commit/096d257624e1ab35076dc89a13046965de362ab5
20160224 05:54:57< irker475> wesnoth: Charles Dang wesnoth:master a33a6698580b / src/quit_confirmation.hpp: Mark quit_confirmation constructor as explicit https://github.com/wesnoth/wesnoth/commit/a33a6698580b7a0ae79b54eb704e1ba94ea646b5
20160224 05:55:12-!- celticminstrel is now known as celmin|sleep
20160224 05:56:22< SigurdFD> is there an easy way to take my PR's for typos on master and make PR's for 1.12?
20160224 05:57:31< vultraz> you could cherry pick them onto the 1.12 branch but that won't work unless the files in question haven't deviated 
20160224 05:57:34< vultraz> (much)
20160224 05:57:50< SigurdFD> I think the files in question haven't deviated at all
20160224 05:58:36< SigurdFD> I'll look up that cherry pick thing.
20160224 06:01:31-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20160224 06:02:39-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has quit []
20160224 06:08:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 06:11:14-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 06:11:15< travis-ci> wesnoth/wesnoth#8588 (master - 1ff15d3 : Charles Dang): The build is still failing.
20160224 06:11:15< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111399262
20160224 06:11:15-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 06:13:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160224 06:55:36-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 06:55:37< travis-ci> wesnoth/wesnoth#8590 (master - a33a669 : Charles Dang): The build has errored.
20160224 06:55:37< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111401445
20160224 06:55:37-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 07:02:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 07:07:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160224 07:09:58-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has quit [Remote host closed the connection]
20160224 07:11:34-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has joined #wesnoth-dev
20160224 07:49:20< irker475> wesnoth: Andreas Löf wesnoth:guifixes d00e20720273 / src/ (display.cpp events.cpp storyscreen/render.cpp storyscreen/render.hpp video.cpp): Make storyscreen respect draw-all events https://github.com/wesnoth/wesnoth/commit/d00e207202735498841be866c48f30c3e8858a03
20160224 07:49:22< irker475> wesnoth: Andreas Löf wesnoth:guifixes 12d1fd006fd3 / data/themes/_initial.cfg src/display.cpp src/display.hpp src/widgets/button.cpp: Resize action buttons dynamically on the main display https://github.com/wesnoth/wesnoth/commit/12d1fd006fd359b66d044d0ad36679dbd8e3c892
20160224 07:49:55< Aginor> vultraz: ^-^ ^^ I think that is the last commit needed
20160224 07:50:22< Aginor> I'll wait for travis, then I'll update the PR so that people can review it properly/again
20160224 07:51:37< vultraz> :D :D
20160224 07:52:58< ancestral> Alright, I’m making a new theme
20160224 07:56:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 08:01:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160224 08:05:42< vultraz> ancestral: ahaha what
20160224 08:05:49< vultraz> Voodo, that is
20160224 08:05:51< vultraz> voodoo 
20160224 08:07:59-!- vn971 [~vasya@91.247.233.240] has left #wesnoth-dev ["ISON Rogach NickServ "]
20160224 08:19:44-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 08:19:45< travis-ci> wesnoth/wesnoth#8591 (guifixes - 12d1fd0 : Andreas Löf): The build was fixed.
20160224 08:19:45< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111415259
20160224 08:19:45-!- travis-ci [~travis-ci@ec2-54-224-204-215.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 08:29:13< vultraz> Aginor: ^ :D
20160224 08:37:19< ancestral> Apparently
20160224 08:37:37< ancestral> I cannot scale images in the theme?
20160224 08:38:40< ancestral> Oh wait I’m doing it wrong
20160224 08:40:12-!- Kwandulin [~Miranda@p200300760F0BC5BCE849E64EDF47759E.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160224 08:43:23< Aginor> looks rather promising
20160224 08:44:48 * Aginor grumbles
20160224 08:44:56< Aginor> now I need to go through the bugtracker
20160224 08:48:11-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160224 08:51:06-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 08:55:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160224 09:01:27 * Aginor swears
20160224 09:01:38< Aginor> the editor is pretty broken in my branch
20160224 09:01:44< Aginor> good thing I thought to test that
20160224 09:01:47 * Aginor sighs
20160224 09:02:41< Aginor> oh well, at least that shouldn't be too painful to fix, I hope
20160224 09:03:42< irker475> wesnoth: Charles Dang wesnoth:guifixes 07169f5e7afe / src/gui/dialogs/ (preferences_dialog.cpp preferences_dialog.hpp): tpreferences: fixup ad8468698916 https://github.com/wesnoth/wesnoth/commit/07169f5e7afe146bd0a02190183bfd7710039f2f
20160224 09:03:45< irker475> wesnoth: Charles Dang wesnoth:guifixes c08a87a255cb / src/gui/dialogs/ (preferences_dialog.cpp preferences_dialog.hpp): tpreferences: hopefully silenced unused parameter warning https://github.com/wesnoth/wesnoth/commit/c08a87a255cb05d2e7cca67708e9b6754455a27e
20160224 09:03:48< irker475> wesnoth: Charles Dang wesnoth:guifixes 81389f4c880f / data/gui/default/widget/toggle_button_listbox_header.cfg: gui2: tweaked layout of listbox header toggle button definition https://github.com/wesnoth/wesnoth/commit/81389f4c880f3945d8cb96b8cee1789debf00563
20160224 09:03:51< irker475> wesnoth: Charles Dang wesnoth:guifixes 34153effa85e / data/gui/default/window/ (game_load.cfg unit_create.cfg): gui2: removed borders from usecases of listbox_header type toggle buttons https://github.com/wesnoth/wesnoth/commit/34153effa85e135984bc84797c54ef2b5c979aff
20160224 09:03:54< irker475> wesnoth: Charles Dang wesnoth:guifixes 710b1fc3afb3 / data/gui/default/window/game_load.cfg: tgame_load: very subtle layout fix https://github.com/wesnoth/wesnoth/commit/710b1fc3afb3f25d5e565e3a9a1fffe0b1f339ee
20160224 09:03:57< irker475> wesnoth: Charles Dang wesnoth:guifixes 67cf95a5af3c / data/gui/default/widget/text_box_default.cfg: gui2: cleanup of the text box widget design https://github.com/wesnoth/wesnoth/commit/67cf95a5af3caefb3123da806126282997c1121d
20160224 09:04:00< irker475> wesnoth: ln-zookeeper wesnoth:guifixes e85d504fbcca / data/core/ (31 files in 3 dirs): Replaced feathered beach wave masks with hard-edged ones https://github.com/wesnoth/wesnoth/commit/e85d504fbccac2aee80a1f24fe42351d5c51ac73
20160224 09:04:02< irker475> wesnoth: Charles Dang wesnoth:guifixes 04d510a91cb9 / data/gui/default/macros/ (horizontal_scrollbar.cfg vertical_scrollbar.cfg): gui2: dropped repeating buttons from scrollbars https://github.com/wesnoth/wesnoth/commit/04d510a91cb9d9ff97a56b09690ca9499c82ab30
20160224 09:04:05< irker475> wesnoth: Charles Dang wesnoth:guifixes 5aa86c42f004 / / (74 files in 3 dirs): Moved scrollbar images to their own folder https://github.com/wesnoth/wesnoth/commit/5aa86c42f004b1da191c645b9d79c5326b21b67e
20160224 09:04:08< irker475> wesnoth: Charles Dang wesnoth:guifixes b57def51302d / data/gui/default/ (widget/label_title_screen_revision.cfg window/title_screen.cfg): ttitle_screen: use a standard label definition for version label https://github.com/wesnoth/wesnoth/commit/b57def51302d70220dfe7ecee5ca3e7254cdcfe8
20160224 09:04:11< irker475> wesnoth: Charles Dang wesnoth:guifixes c0254f6bcae1 / src/gui/widgets/control.cpp: tcontrol: add text_alignment setter to set_members https://github.com/wesnoth/wesnoth/commit/c0254f6bcae1b4174a0a4bd5e3e39d1fcc11b7bb
20160224 09:04:14< irker475> wesnoth: Charles Dang wesnoth:guifixes e8c3fdb2381c / / (2 files in 2 dirs): tcampaign_dialog: add support for setting alignment of description text from WML https://github.com/wesnoth/wesnoth/commit/e8c3fdb2381c2b2cf4dc83b3e7583268c8e36f0b
20160224 09:04:17< irker475> wesnoth: Charles Dang wesnoth:guifixes 9cdc3bcbd2a3 / data/campaigns/Sceptre_of_Fire/_main.cfg: SoF: center description poem https://github.com/wesnoth/wesnoth/commit/9cdc3bcbd2a39c78fc2515af24810283763cb431
20160224 09:04:20< irker475> wesnoth: Charles Dang wesnoth:guifixes 499cadc37b99 / src/widgets/scrollbar.cpp: Update GUI1 scrollbar image paths post-5aa86c42f004 https://github.com/wesnoth/wesnoth/commit/499cadc37b99934ec8352a337a5f55292efe0775
20160224 09:04:23< irker475> wesnoth: gfgtdf wesnoth:guifixes cc5dc5b382ea / src/game_initialization/playcampaign.cpp: attempt to fix replays of mp games not beeing loaded correctly. https://github.com/wesnoth/wesnoth/commit/cc5dc5b382ea74aaff291de1fdc6e3e888c0f711
20160224 09:04:25< irker475> wesnoth: Charles Dang wesnoth:guifixes 847ee990a31b / data/gui/default/window/campaign_dialog.cfg: Fixup e8c3fdb2381c https://github.com/wesnoth/wesnoth/commit/847ee990a31b88673c39df1f1ec1f0101dfd11a2
20160224 09:04:28< irker475> wesnoth: Charles Dang wesnoth:guifixes 2b6566682386 / src/gui/dialogs/preferences_dialog.cpp: tpreferences: ensure all descriptions are shown https://github.com/wesnoth/wesnoth/commit/2b6566682386cb4e30d474765e8b5d3f8b26cab3
20160224 09:04:31< irker475> wesnoth: Ignacio R. Morelle wesnoth:guifixes 74973218b150 / data/gui/default/macros/ (horizontal_scrollbar.cfg vertical_scrollbar.cfg): Revert "gui2: dropped repeating buttons from scrollbars" https://github.com/wesnoth/wesnoth/commit/74973218b1502c78efc2cedfa256a5e0fdba25dc
20160224 09:04:34< irker475> wesnoth: Charles Dang wesnoth:guifixes 2d9996093010 / data/campaigns/Under_the_Burning_Suns/scenarios/02_Across_the_Harsh_Sands.cfg: UtBS S2: fixed Go’hag not being included in undead group death event https://github.com/wesnoth/wesnoth/commit/2d99960930106464b3cd12517fe7fcd57d10d4df
20160224 09:04:37< irker475> wesnoth: Charles Dang wesnoth:guifixes 7f02961ddaf4 / data/campaigns/Under_the_Burning_Suns/scenarios/02_Across_the_Harsh_Sands.cfg: Fixup 2d9996093010 https://github.com/wesnoth/wesnoth/commit/7f02961ddaf467a29882a4acd32aa55201188ad5
20160224 09:04:40< irker475> wesnoth: gfgtdf wesnoth:guifixes a5d03e951995 / src/unit_filter.cpp: remove support for name= in standart unit filters https://github.com/wesnoth/wesnoth/commit/a5d03e9519954bf09ca17f52fe51438146bafdb7
20160224 09:04:42< irker475> wesnoth: gfgtdf wesnoth:guifixes 89e8341b1aa1 / data/campaigns/Under_the_Burning_Suns/scenarios/02_Across_the_Harsh_Sands.cfg: fix use of invalid name= attribute in standard unit filtes in UtbS scenario2 https://github.com/wesnoth/wesnoth/commit/89e8341b1aa1ff72a05c014c1136b481df9479cd
20160224 09:04:45< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes 43e3f8e63164 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: XCode: Split up copy files phase a little https://github.com/wesnoth/wesnoth/commit/43e3f8e6316462fe1f3d837ad63d073ac0ca62a9
20160224 09:04:46< irker475> wesnoth: Charles Dang wesnoth:guifixes 1a6bfca90356 / data/gui/default/window/preferences/ (01_general.cfg 02_display.cfg 03_sound.cfg 05_advanced.cfg): tpreferences: cleaned up some unnecessary grid nesting https://github.com/wesnoth/wesnoth/commit/1a6bfca90356f90a182ae4267f16ad11eaa9d4c7
20160224 09:04:49< irker475> wesnoth: Charles Dang wesnoth:guifixes 05411b6d34ad / data/gui/default/window/preferences/01_general.cfg: tpreferences: made sure Scroll Speed slider fills all available space https://github.com/wesnoth/wesnoth/commit/05411b6d34ad19440565feea8a4bf6d5271bb0a9
20160224 09:04:52< irker475> wesnoth: Charles Dang wesnoth:guifixes f23bc20e8217 / src/gui/dialogs/preferences_dialog.cpp: tpreferences: fixed incorrect use of doxygen comments https://github.com/wesnoth/wesnoth/commit/f23bc20e8217d558cd11f450d9a1633555a6bd11
20160224 09:04:55< irker475> wesnoth: Charles Dang wesnoth:guifixes 168fce195844 / data/campaigns/Legend_of_Wesmere/scenarios/chapter1/03_Kalian_under_Attack.cfg: LoW S3: fixed incorrectly placed labels (bugs #22931, #23776) https://github.com/wesnoth/wesnoth/commit/168fce195844c0b6b840373da240fdee90b1590e
20160224 09:04:58< irker475> wesnoth: Charles Dang wesnoth:guifixes cca804e78d3b / src/team.cpp: Add 'is_local' to recognized key list https://github.com/wesnoth/wesnoth/commit/cca804e78d3b08e6c940bb30ef43ccb5ae41142f
20160224 09:05:02< irker475> wesnoth: Charles Dang wesnoth:guifixes 1c0085616a8a / src/ (hotkey_handler_sp.cpp playsingle_controller.hpp): Implement is_replay() in playsingle_controller https://github.com/wesnoth/wesnoth/commit/1c0085616a8a07da937d4bc636b5f496c64f8a44
20160224 09:05:05< irker475> wesnoth: Charles Dang wesnoth:guifixes 800a590525be / src/synced_context.cpp: Moved synced_context replay check deeper https://github.com/wesnoth/wesnoth/commit/800a590525bea30305fe68afe152ffa1eb571ad4
20160224 09:05:07< irker475> wesnoth: Charles Dang wesnoth:guifixes 53190dfb4c55 / src/replay.cpp: Fixed a typo in an error https://github.com/wesnoth/wesnoth/commit/53190dfb4c550da77266f6d673e4af14eb5f6322
20160224 09:05:11< irker475> wesnoth: gfgtdf wesnoth:guifixes 3683490ac688 / src/unit_filter.cpp: add wanring for name= in standard unit filters. https://github.com/wesnoth/wesnoth/commit/3683490ac6885e58c7577a3080e747975dd1f9dc
20160224 09:05:13< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes fd939cee76c4 / src/unit_filter.cpp: Unit filters: move name warning to constructor and suggest alternatives https://github.com/wesnoth/wesnoth/commit/fd939cee76c47207db73f71a8194a3d2611f7252
20160224 09:05:14< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes 18bcd519e0dd / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: XCode: Suppress warnings about unknown warnings https://github.com/wesnoth/wesnoth/commit/18bcd519e0dd9de9f411e0bb1aecd77f555629eb
20160224 09:05:17< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes 708af5640953 / src/scripting/ (game_lua_kernel.cpp lua_api.cpp lua_api.hpp): Fix function name typo https://github.com/wesnoth/wesnoth/commit/708af564095390490048d04fbd40709cbcda9185
20160224 09:05:18< irker475> wesnoth: Robert Wallace wesnoth:guifixes afd53cf6a1b2 / data/multiplayer/scenarios/ANL_utils/ (4 files): Fixed deprecated code in ANL https://github.com/wesnoth/wesnoth/commit/afd53cf6a1b20b9f5efc6143274faa888a308a3f
20160224 09:05:20< irker475> wesnoth: Robert Wallace wesnoth:guifixes df45dd9aab03 / data/core/about.cfg: Update about.cfg https://github.com/wesnoth/wesnoth/commit/df45dd9aab034ff08c0af27ac087bf986dddc603
20160224 09:05:23< irker475> wesnoth: gfgtdf wesnoth:guifixes 0484fdd4149f / src/ (9 files in 4 dirs): fix quit confirmation promt in editor https://github.com/wesnoth/wesnoth/commit/0484fdd4149fc794d59517f7a19873bbb5fe6ad6
20160224 09:05:25< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes baed863df073 / src/ (27 files in 10 dirs): Don't include config.hpp when possible https://github.com/wesnoth/wesnoth/commit/baed863df07360f8753c81d160c3ea202da681c5
20160224 09:05:27< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes c08ea95c7256 / src/config.hpp: config: make static string constants private https://github.com/wesnoth/wesnoth/commit/c08ea95c72566614a5d44a347e2517f4491d0e70
20160224 09:05:29< irker475> wesnoth: Charles Dang wesnoth:guifixes c0df87a47049 / src/help/help_impl.hpp: Restore a config.hpp include https://github.com/wesnoth/wesnoth/commit/c0df87a4704924ba8331858e595090a74da1a638
20160224 09:05:32< irker475> wesnoth: Charles Dang wesnoth:guifixes 56f0c9a99a6d / src/help/help_impl.hpp: Redid c0df87a47049 correctly https://github.com/wesnoth/wesnoth/commit/56f0c9a99a6d42359db2407acc367f6e5f5863ee
20160224 09:05:34< irker475> wesnoth: Charles Dang wesnoth:guifixes c40c2cd9af52 / data/multiplayer/scenarios/ANL_utils/ (ANL_research_options.cfg ANL_worker_options.cfg): ANL: strip unnecessary wmltools guards and ran wmlindent https://github.com/wesnoth/wesnoth/commit/c40c2cd9af52d1dbfecbad38684b92ce3dd5ba8b
20160224 09:05:38< irker475> wesnoth: Charles Dang wesnoth:guifixes 2346b3508f8d / src/hotkey/ (command_executor.cpp command_executor.hpp): Simplify handling of HOTKEY_QUIT_TO_DESKTOP https://github.com/wesnoth/wesnoth/commit/2346b3508f8d58b0581ad0485b63014094faea67
20160224 09:05:41< irker475> wesnoth: Charles Dang wesnoth:guifixes b73cc5eb08f4 / src/ (quit_confirmation.cpp quit_confirmation.hpp): Small formatting cleanup https://github.com/wesnoth/wesnoth/commit/b73cc5eb08f4a26f4e7284c1236fe4dc73bb2d22
20160224 09:05:44< irker475> wesnoth: Celtic Minstrel wesnoth:guifixes f067332d0ff5 / src/ (4 files in 3 dirs): fix spelling https://github.com/wesnoth/wesnoth/commit/f067332d0ff5c78541f5cb31c4cdff47f83e61cb
20160224 09:05:46< irker475> wesnoth: Charles Dang wesnoth:guifixes ed9ead2ac9e4 / src/ (6 files in 3 dirs): Add flag to quit_confirmation::quit() to only quit out of game https://github.com/wesnoth/wesnoth/commit/ed9ead2ac9e414efe52cab34f62fe2652cd88270
20160224 09:05:49< irker475> wesnoth: Charles Dang wesnoth:guifixes 27d4091ec5bf / src/quit_confirmation.hpp: Reapplied a fix from f067332d0ff5 that got lost in ed9ead2ac9e4 https://github.com/wesnoth/wesnoth/commit/27d4091ec5bf510a9f3a1dd2a50d46ac229cfb8c
20160224 09:05:52< irker475> wesnoth: Charles Dang wesnoth:guifixes 7d0a02ec7d1a / src/ (7 files in 4 dirs): Bunch of refactoring of editor quit handling https://github.com/wesnoth/wesnoth/commit/7d0a02ec7d1af805e175f58d4adcbf3faef1b8ed
20160224 09:05:55< irker475> wesnoth: Charles Dang wesnoth:guifixes 03ed7ffdff24 / src/editor/ (controller/editor_controller.cpp map/context_manager.cpp): Applied some formatting to Editor quit confirmation-with-maps-open message https://github.com/wesnoth/wesnoth/commit/03ed7ffdff24f5b9cfe7327b3b5b389c18cbb3dd
20160224 09:05:58< irker475> wesnoth: Charles Dang wesnoth:guifixes 031628dac5bf / src/editor/controller/editor_controller.cpp: Small Editor quit confirmation message wording change https://github.com/wesnoth/wesnoth/commit/031628dac5bfe95976e599735c2cd725e5ea095d
20160224 09:06:01< irker475> wesnoth: sigurdfdragon wesnoth:guifixes 7d2eaef38d4f / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): AOI: Make the named loyal units unnameable. https://github.com/wesnoth/wesnoth/commit/7d2eaef38d4fed1f10bdb2aad01ed2163f794270
20160224 09:06:03< irker475> wesnoth: Charles Dang wesnoth:guifixes c9e2dd19db5b / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): Merge pull request #607 from sigurdfdragon/AOI_Named_Loyals https://github.com/wesnoth/wesnoth/commit/c9e2dd19db5b044a252708efbf3b6c6a97c1a3a1
20160224 09:06:06< irker475> wesnoth: Charles Dang wesnoth:guifixes a87939acb02d / data/multiplayer/scenarios/ (2p_Dark_Forecast.cfg Random_Scenario_Lua_Cave.cfg): Run wmlindent https://github.com/wesnoth/wesnoth/commit/a87939acb02df8f6e3fa2ba626449a3574930963
20160224 09:06:09< irker475> wesnoth: Charles Dang wesnoth:guifixes 1ff15d3ff168 / data/campaigns/ (22 files in 11 dirs): Ran wmlindent on campaigns https://github.com/wesnoth/wesnoth/commit/1ff15d3ff1680ca55e71741fde95b379fbf3eb25
20160224 09:06:12< irker475> wesnoth: Charles Dang wesnoth:guifixes 5418ddfc2495 / src/gui/dialogs/ (synced_choice_wait.cpp synced_choice_wait.hpp): tsynced_choice_wait: removed now-unused CVideo parameter from pre_show https://github.com/wesnoth/wesnoth/commit/5418ddfc24955de2e83addb150b9dd6848bdfb71
20160224 09:06:15< irker475> wesnoth: Charles Dang wesnoth:guifixes 096d257624e1 / src/gui/dialogs/ (synced_choice_wait.cpp synced_choice_wait.hpp): Redo 5418ddfc2495 properly https://github.com/wesnoth/wesnoth/commit/096d257624e1ab35076dc89a13046965de362ab5
20160224 09:06:18< irker475> wesnoth: Charles Dang wesnoth:guifixes a33a6698580b / src/quit_confirmation.hpp: Mark quit_confirmation constructor as explicit https://github.com/wesnoth/wesnoth/commit/a33a6698580b7a0ae79b54eb704e1ba94ea646b5
20160224 09:06:21< irker475> wesnoth: Andreas Löf wesnoth:guifixes c0f7fb6be25e / / (206 files in 39 dirs): Merge remote-tracking branch 'origin/master' into guifixes https://github.com/wesnoth/wesnoth/commit/c0f7fb6be25e9d10ca054d6ea00cc703d520e5b2
20160224 09:24:51-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160224 09:30:26< zookeeper> Aginor, random question: very sleepy shows SDL_LogCritical taking up an awful lot of CPU time. do you happen to know anything relating to that?
20160224 09:31:21< zookeeper> because i've tried to look into it to verify whether it's just a glitch of some kind or if it's actually a genuine problem (maybe only with debug-enabled builds), but to no avail.
20160224 09:37:05< Aginor> zookeeper: sorry, no, I have no idea what might cause it
20160224 09:37:13< zookeeper> all right
20160224 09:37:23< Aginor> I don't think I have an SDL-debug build though
20160224 09:37:42< zookeeper> the SDL2 sources are too obfuscated for me to figure out where in its internals it can get called from
20160224 09:38:20< zookeeper> because the traces don't show it coming directly from any of the wesnoth calls, but maybe that's just because SDL hasn't been debug-compiled or whatever. i dunno. it still bugs me.
20160224 09:38:30< Aginor> yup
20160224 09:38:50< Aginor> is there any way for you to share the results with me?
20160224 09:39:12< Aginor> question for the channel: how do I debug the following error: 20160224 22:37:38 error display: could not open image 'icons/action/editor-tool-unit.png'
20160224 09:39:30< Aginor> and should it have a size embedded in it?
20160224 09:39:44< Aginor> I'm also seeing 20160224 22:37:38 error display: could not open image 'icons/action/editor-switch-time_30_30-pressed.png'
20160224 09:39:50< zookeeper> Aginor, well sure i can show you some screenshots showing it
20160224 09:40:03< Aginor> where it seems odd that the size is there twice
20160224 09:40:12< Aginor> zookeeper: yes please
20160224 09:40:26< Aginor> zookeeper: I don't think I can be of any immediate help, but I am interested in your findings
20160224 09:42:50< vultraz> Aginor: means image paths are wrong
20160224 09:42:54< vultraz> first has no size
20160224 09:43:00< vultraz> second has one size too manu
20160224 09:43:12< vultraz> many*
20160224 09:43:33< Aginor> vultraz: what's computing/altering them though? Somewhere there's code that mucks about with them because it doesn't stack up with the WML
20160224 09:44:58< zookeeper> Aginor, https://dl.dropboxusercontent.com/u/63964618/wesnoth/sdl_logcritical_sleepy.png
20160224 09:45:13-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 09:45:33< zookeeper> maybe all the actual low-level draw ops are just wrapped in those kind of logging calls or something
20160224 09:45:43< zookeeper> it would make the most sense
20160224 09:45:45< Aginor> that really is a lot of cpu-time by comparison
20160224 09:45:51< Aginor> let me check the source
20160224 09:46:00 * Aginor happens to have lots of sdl sources around
20160224 09:46:00< zookeeper> good luck :P
20160224 09:46:12< Aginor> I have spent a lot of time digging through the SDL code
20160224 09:46:40< zookeeper> it's the kind of boilerplate c that i can't decipher
20160224 09:47:31< Aginor> that turned nasty quickly
20160224 09:47:47< zookeeper> there's other stacktraces there that i can browse (with the last identifiable calls being wesnoth calls calling some SDL functions), but they account for <1%
20160224 09:48:15< zookeeper> googling hasn't turned up anything much, nor did asking on #sdl
20160224 09:49:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160224 09:50:08< Aginor> I don't think LogCritical is a part of the internals, I rather think it's exposed functionality
20160224 09:50:42< Aginor>  *  Simple log messages with categories and priorities.
20160224 09:50:42< Aginor>  *
20160224 09:50:42< Aginor>  *  By default logs are quiet, but if you're debugging SDL you might want:
20160224 09:50:42< Aginor>  *
20160224 09:50:42< Aginor>  *      SDL_LogSetAllPriority(SDL_LOG_PRIORITY_WARN);
20160224 09:50:45< Aginor>  *
20160224 09:50:47< Aginor>  *  Here's where the messages go on different platforms:
20160224 09:50:50< Aginor>  *      Windows: debug output stream
20160224 09:50:52< Aginor>  *      Android: log output
20160224 09:50:55< Aginor>  *      Others: standard error output (stderr)
20160224 09:50:58< Aginor> zookeeper: are you finding anything in the debug output stream?
20160224 09:51:16< zookeeper> nope
20160224 09:51:33< zookeeper> assuming by that you just mean the logs
20160224 09:51:44< Aginor> that's from SDL_log.h
20160224 09:51:56< zookeeper> oh right. i have no idea what debug output stream is.
20160224 09:52:03< zookeeper> vultraz, whether someone went through the trouble of naming a unit doesn't really have any relevance to whether it should be renamable by the player if they want to. the only thing that really matters is whether the name is used in dialogue or other text.
20160224 09:52:04< Aginor> neither do I
20160224 09:52:07< Aginor> it's a windows thing
20160224 09:52:28< vultraz> zookeeper: I disagree
20160224 09:53:16< Aginor> zookeeper: I have no idea, sorry
20160224 09:53:29< Aginor> and it's getting bed-time late for me
20160224 09:53:32< zookeeper> Aginor, okay, i'll see if i can find out what it is
20160224 09:53:43< zookeeper> vultraz, well of course you do. when do you not? :p
20160224 09:54:06< vultraz> zookeeper: when you're improving performance of the water graphics
20160224 09:56:31< zookeeper> anyway, even if it's an incredibly minor change it's still wrong.
20160224 09:57:15< vultraz> no it's not
20160224 09:57:39< vultraz> why are you of the opinion that the player should get to do everything
20160224 09:58:17< zookeeper> because it's their spare time and they get to decide how they want to use it
20160224 09:58:40< vultraz> nope
20160224 09:58:59< vultraz> we're creating an experience and get to dictate how it's experienced 
20160224 09:59:11< zookeeper> no we're not, that's just you
20160224 09:59:58< vultraz> why do I think you'd be a Republican :P 
20160224 10:00:39< zookeeper> that's a very strange conclusion, even from you
20160224 10:01:03< vultraz> Republicans are for fewer government restrictions 
20160224 10:01:57< vultraz> You're a proponent of little control over the experience 
20160224 10:03:08< zookeeper> maybe
20160224 10:04:35< vultraz> Whereas I prefer the 'we decide what the players get to do' mentality 
20160224 10:05:05< zookeeper> i know
20160224 10:06:19< vultraz> (and I'm a Democrat :D )
20160224 10:08:07< iceiceice> vultraz, high five
20160224 10:08:21< iceiceice> lol
20160224 10:08:47< vultraz> was that sarcastic? :P
20160224 10:08:52< iceiceice> no, i seriously am a democrat
20160224 10:08:59< iceiceice> i'm always happy when people proudly identify as such
20160224 10:09:04< iceiceice> i grew up in an ultra-conservative town
20160224 10:09:05< iceiceice> hehe
20160224 10:09:29< vultraz> ahh
20160224 10:09:31< vultraz> :D
20160224 10:09:37< iceiceice> that being said, i can't say i'm entirely impressed with the debate tactic of "compare your opponents to the republicans"
20160224 10:09:45< iceiceice> which is honestly, a close cousin of, compare your opponents to hitler
20160224 10:09:54< iceiceice> :p
20160224 10:10:27< iceiceice> see also godwin's law
20160224 10:10:57< vultraz> it doesn't really apply anyway since I think zookeeper's finnish
20160224 10:11:17< zookeeper> US politics are something that any civilized part of the world would collectively facepalm at, constantly. which is about as much as i'm interested in saying about the subject in this context :p
20160224 10:11:44< iceiceice> yeah US politics has gotten so wierd at this point
20160224 10:11:54< iceiceice> its just boringly wierd
20160224 10:22:35< zookeeper> vultraz, anyway, i'm for minimal restriction because experience has taught me that that's the path of more fun in a game. as a creator it's a very appealing idea to try to restrict the player into experiencing what you want them to experience the way you intended it, but in practise, as a player it is almost never nothing but annoying when you hit those barriers. it never feels like an 
20160224 10:22:36< zookeeper> invisible hand guiding you towards an enjoyable experience, it always feels like an invisible wall put there to prevent you from doing something you wanted to do.
20160224 10:23:29-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 240 seconds]
20160224 10:23:42< zookeeper> hence i try to curb by instincts to try to force the player to do what i want them to do, and just let them do what they want to do, because the former simply will not make them have more fun.
20160224 10:23:46< zookeeper> s/by/my
20160224 10:24:29< vultraz> finally, a convincing argument
20160224 10:26:53< zookeeper> and i'm sorry if this sounds patronizing, but i believe you're still in the phase where you think that you can guide the player towards a more enjoyable experience by restrictions, instead of incentives
20160224 10:27:44< vultraz> im in a dota match r/n
20160224 10:27:46< vultraz> will get back to you
20160224 10:31:22< zookeeper> and by restrictions i mean the game preventing you from doing what it usually lets you do, with the obvious intent of trying to prevent you from doing something that the game thinks wouldn't be very fun (like naming a unit "smellypants").
20160224 10:32:07< vultraz> I agree you can do that
20160224 10:32:09< zookeeper> which completely ignores the issue that discovering things to do is what makes a game fun, even if it's just choosing to give one of your elvish rangers a funny name.
20160224 10:32:11< vultraz> but not main characters
20160224 10:32:25< zookeeper> none of those units are main characters.
20160224 10:35:58< zookeeper> none of them will ever say a single word after getting (potentially) renamed by the player, except one line by a random advisor in the victory event, and if the ranger you named "smellypants" happens to get that line then i can guarantee the player won't feel disappointed by that.
20160224 10:36:59< zookeeper> in fact that would probably be the only fun memorable thing to take away from it :p
20160224 10:39:21-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 10:44:02-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has joined #wesnoth-dev
20160224 10:44:05-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160224 10:44:46< Elvish_Hunter> Hi all :)
20160224 10:45:05< Elvish_Hunter> So, I read the logs about the [filter] name= issue.
20160224 10:45:16< zookeeper> _if_ it was possible to allow renaming of main characters then i'm very much think that it should be allowed. but i don't think that's something you need to worry about ever happening.
20160224 10:45:45< Elvish_Hunter> My question is: does anyone think that it might be useful to add a new wmllint rule that warns if name= is found in [filter]?
20160224 10:46:16< Ravana_> it is pointless to disable renaming units when player can just get [modification] to get around the limit
20160224 10:46:28-!- mjs-de [~mjs-de@f048076100.adsl.alicedsl.de] has joined #wesnoth-dev
20160224 10:47:46 * zookeeper sees no logic in that
20160224 10:47:51< Elvish_Hunter> Ravana_: that may be considered a bug in [modifications] then.
20160224 10:48:45< Ravana_> restricting WML seems even worse idea
20160224 10:51:05< Ravana_> say you make wml unable to rename unit - that also means you need to make wml unable to create units, else just new unit with everything from name taken from previous unit
20160224 10:53:32< vultraz> zookeeper: so to sum up, you're advocating the 'easter egg' principle?
20160224 10:54:05< zookeeper> what principle is that? being able to rename units is hardly an easter egg in itself
20160224 10:54:30< vultraz> keeping small things in the game for the player to discover and have fun with
20160224 10:54:54< zookeeper> yes, of course.
20160224 10:55:00< vultraz> that are sometimes enjoyable just by the merit of being easter eggs
20160224 10:56:12< iceiceice> i think the game engine should be simple and predictable
20160224 10:56:17< Elvish_Hunter> But anyway, my question was about the usefulness of a wmllint rule.
20160224 10:56:18< iceiceice> the best tools are tools that don't surprise you
20160224 10:56:23< iceiceice> that's my 2 cents
20160224 10:56:25< iceiceice> not that anyone cares
20160224 10:56:49< iceiceice> note that doesnt mean that filtering on name is bad necessarily,
20160224 10:57:06< iceiceice> if you can do it in a way that is simple and predictable then its probably good
20160224 10:57:30< vultraz> zookeeper: alright, if you want to revert the commit you can
20160224 10:57:34< iceiceice> but my tentative assumption is that that is hard, or at least the engine doesn't do that
20160224 10:57:38< vultraz> zookeeper: but you'll have to explain to SFD 
20160224 11:00:02< zookeeper> iceiceice, well it's not hard, the filter is like a one-liner. possible complications due to existing multiplayer bugs are another thing of course, but that basically has nothing to do with the name filter implementation.
20160224 11:00:15< iceiceice> yeah but it relies on like
20160224 11:00:22< iceiceice> potentially converting strings to tstrings
20160224 11:00:27< iceiceice> or back
20160224 11:00:41< iceiceice> i mean tstrings are generally pretty complicated
20160224 11:01:04< iceiceice> and i somewhat doubt that they were meant to be compared with one-another
20160224 11:01:27< vultraz> iceiceice: whatever happened to your lightweight boost::spirit-based dynamic translation loader
20160224 11:01:32< iceiceice> it exists
20160224 11:01:38< iceiceice> some people starred it on github :)
20160224 11:01:46< iceiceice> they offered to put a link to it on the boost spirit website
20160224 11:01:53< iceiceice> i think i will take them up on that later
20160224 11:01:58< iceiceice> vultraz, but it requires C++!1
20160224 11:02:00< iceiceice> *C++11
20160224 11:02:12< vultraz> ah...
20160224 11:02:13< vultraz> yes...
20160224 11:02:20< vultraz> the elusive c++ standard from 5 years ago 
20160224 11:02:39< iceiceice> it could potentially be rewritten to support C++98
20160224 11:02:47< iceiceice> but if you do it naively it probably won't be terribly efficient
20160224 11:02:50< vultraz> I hope we'll make the switch to c++11 in the next few months
20160224 11:02:51< iceiceice> what with no move semantics at all
20160224 11:02:56< iceiceice> idk
20160224 11:02:59< iceiceice> it is only 700 lines
20160224 11:03:08< iceiceice> so if someone wanted to port it to C++98 thats probably quite doable
20160224 11:03:12< vultraz> 700??? O_O that small??
20160224 11:03:13< iceiceice> :)
20160224 11:03:21< iceiceice> boost spirit is a super powerful library
20160224 11:03:22< vultraz> that's smaller than the entire tpreferences class
20160224 11:03:34< iceiceice> it is really pretty amazing
20160224 11:03:56< vultraz> we should look into using it..
20160224 11:04:02< iceiceice> that would be cool :)
20160224 11:04:03< vultraz> once we get 11
20160224 11:04:10< iceiceice> i made a very rigorous test suite actually
20160224 11:04:16< iceiceice> i basically downloaded about 100 assorted .po files
20160224 11:04:20< iceiceice> from the Gnome project
20160224 11:04:23< iceiceice> some from wesnoth
20160224 11:04:26< iceiceice> some from frogatto actually
20160224 11:04:35< iceiceice> i think i got at least 2/3 of the languages in the gnome project
20160224 11:04:41< vultraz> I think gfgtdf and zookeeper are the last people who need to be able to build with c++11
20160224 11:04:52< iceiceice> including the languages with astounding numbers of plural forms
20160224 11:05:19< iceiceice> actually there are still some significant ways it could be improved i guess
20160224 11:05:23< iceiceice> hehe
20160224 11:05:37< iceiceice> if someone wants to make a byte-code for the plural forms evaluation engine
20160224 11:05:53-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160224 11:05:54< iceiceice> that would probably make a signifiacnt difference for the speed at which we can translate arabic i guess
20160224 11:06:02< iceiceice> although its probably negligible in any real application
20160224 11:06:42-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160224 11:07:02< vultraz> how much a performance gain would it give wesnoth?
20160224 11:07:26< iceiceice> i dont think that, converting a number "n" to select the plural form index for any given language is actually a bottle neck in any program
20160224 11:07:33< iceiceice> except like some benchmark program specifically for that purpose
20160224 11:07:42< iceiceice> do you mean like , how much gain would it give wesnoth generally?
20160224 11:07:59< vultraz> yes
20160224 11:08:02< iceiceice> vultraz, the main thign that it does is, it makes it so that you dont need to have wescamp
20160224 11:08:15< iceiceice> in the current system,
20160224 11:08:26< iceiceice> we make the `.pot` files
20160224 11:08:36< iceiceice> they turn into `.po` files which are filled in by translators
20160224 11:08:39< iceiceice> and become part of the tree
20160224 11:08:48< iceiceice> then, the scons/ cmake scripts are supposed to convert the `.po` to `.mo`
20160224 11:08:51< iceiceice> at the time that you compile wesnoth
20160224 11:09:03< iceiceice> the mo files are like, this fairly goofy optmization
20160224 11:09:13< iceiceice> instead of just having a plaintext dictionary file,
20160224 11:09:23< iceiceice> GNU poeple wanted to have like, some hyper optimized binary format
20160224 11:09:40< iceiceice> mo is not a human readable format, its like a binary sepcification of a hash table
20160224 11:09:48< iceiceice> and its different for every architecture
20160224 11:09:58< iceiceice> and wesnoth can only read `.mo` files 
20160224 11:10:02< iceiceice> because it uses GNU libintl
20160224 11:10:06< iceiceice> and that's what GNU libintl reads
20160224 11:10:15< iceiceice> so basically there's this portability problem
20160224 11:10:21< iceiceice> if i have a campaign
20160224 11:10:31< iceiceice> and i want to send it to the addon server, with translations attached,
20160224 11:10:43< iceiceice> that means that, there needs to be an mo file for linux, an mo file for OS X,
20160224 11:10:55< iceiceice> idk, i mean, for any architecture where there are differences, i need to have a different mo file
20160224 11:11:05< iceiceice> and realisitcally none of the users can actually produce that
20160224 11:11:20< iceiceice> so the solution was this wescamp thing
20160224 11:11:31< iceiceice> where AI0867 runs a shellscript that produces all these mo files at once in a batch
20160224 11:11:41< iceiceice> but that's also fairly tedious and not really a solution
20160224 11:11:43< Elvish_Hunter> iceiceice: actually, a .mo file works on both Windows and Linux.
20160224 11:11:47< iceiceice> yeah you're right
20160224 11:11:50< iceiceice> its the chip, not the OS
20160224 11:11:53< vultraz> this sounds like it could have been avoided had we just supported plain .po files
20160224 11:11:57< iceiceice> yes!
20160224 11:12:00< iceiceice> but there are no libs for that
20160224 11:12:02< iceiceice> so i made one
20160224 11:12:08< vultraz> ahhh
20160224 11:12:08< iceiceice> thats what my lib does
20160224 11:12:13< vultraz> we definitely need to look into this, then
20160224 11:12:28< Elvish_Hunter> Yeah, big endian vs little endian architectures.
20160224 11:12:48< iceiceice> Elvish_Hunter, potentially also like, 32 vs 64 bit
20160224 11:12:55< iceiceice> i dont actually know all the details
20160224 11:13:13< iceiceice> i dont think that like, GNU folks were wrong in 1980 or whenever to design it this way
20160224 11:13:23< iceiceice> because like, libintl needs to work also on all the GNU binutils like sed
20160224 11:13:42< iceiceice> where the cost of parsing the po file, instead of the mo file, may be comparable to the actual run time of sed
20160224 11:13:53< iceiceice> but for like, any normal user application
20160224 11:13:56< iceiceice> on modern hardware
20160224 11:14:07< iceiceice> i fail to see any reason to use mo files
20160224 11:14:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160224 11:15:47< zookeeper> vultraz, here's a better deal: you revert, and i'll explain to SDF :p
20160224 11:17:37< irker475> wesnoth: Charles Dang wesnoth:revert-607-AOI_Named_Loyals aba935060aea / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): Revert "AOI: Make the named loyal units unnameable." https://github.com/wesnoth/wesnoth/commit/aba935060aeaa34d5efe9e2afed5cc4d2f343ff8
20160224 11:17:59< iceiceice> vultraz, it is also a "header only" lib
20160224 11:18:07< iceiceice> so it means that people can stop linking with libintl
20160224 11:18:08< irker475> wesnoth: Charles Dang wesnoth:master aba935060aea / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): Revert "AOI: Make the named loyal units unnameable." https://github.com/wesnoth/wesnoth/commit/aba935060aeaa34d5efe9e2afed5cc4d2f343ff8
20160224 11:18:11< irker475> wesnoth: Charles Dang wesnoth:master b76813c8ac0c / data/campaigns/An_Orcish_Incursion/scenarios/ (01_Defend_the_Forest.cfg 07_Showdown.cfg): Merge pull request #608 from wesnoth/revert-607-AOI_Named_Loyals https://github.com/wesnoth/wesnoth/commit/b76813c8ac0ccb3144a05d99ea8f4b99d1f3d2de
20160224 11:18:18< iceiceice> and the gettext tools are no longer a build-requirement i guess
20160224 11:18:23< iceiceice> if you switch to it
20160224 11:18:26< vultraz> zookeeper: I'm a lazy ass who used the web interface :P
20160224 11:19:36< vultraz> iceiceice: t'is good. we'd have to add spirit as a boost requirement but that's not a problem
20160224 11:19:52< iceiceice> no, because spirit is also header only :)
20160224 11:20:06< vultraz> oo
20160224 11:20:06< iceiceice> its like a no-dependencies lib
20160224 11:20:14< iceiceice> as long as you have boost headers, you get it
20160224 11:20:28< vultraz> even better
20160224 11:21:01-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20160224 11:28:44-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160224 11:30:03-!- TC01 [~quassel@london.acm.jhu.edu] has quit [Ping timeout: 240 seconds]
20160224 11:30:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160224 11:33:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 11:37:55-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160224 11:40:58-!- travis-ci [~travis-ci@ec2-54-167-161-251.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 11:40:59< travis-ci> wesnoth/wesnoth#8594 (revert-607-AOI_Named_Loyals - aba9350 : Charles Dang): The build has errored.
20160224 11:40:59< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111450913
20160224 11:40:59-!- travis-ci [~travis-ci@ec2-54-167-161-251.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 11:44:45< fabi> vultraz: The Forum's experimental corner vanished. Were did all the discussions go?
20160224 11:45:20< fabi> s/were/where
20160224 11:46:29< vultraz> dunno
20160224 11:46:50< fabi> I guess that happened without any public discussion...
20160224 11:47:13< vultraz> what even was in there
20160224 11:47:16< vultraz> I can't remeber
20160224 11:48:35< Elvish_Hunter> It was a forum about an experimental fork of Wesnoth. If I remember correctly, Yogibear and fabi worked on it for some time.
20160224 11:49:30< fabi> It was more about all kind of discussion about new features the wesnoth developers weren't able or willing to implement inhouse.
20160224 11:50:37< fabi> I felt the content was very valuable for getting used to what neglected groups like the ladder people need.
20160224 11:51:15< fabi> But it does not surprise me very much that it vanished without any fuzz. The new regime just stinks.
20160224 11:52:02< vultraz> oh for god's sake 
20160224 11:52:07< vultraz> let's not start this argument again
20160224 11:53:11< fabi> vultraz, aren't you communication manager whatever nowadays?
20160224 11:53:18< vultraz> yes
20160224 11:53:31< fabi> I see.
20160224 11:53:58< zookeeper> when was the last time it was visible?
20160224 11:54:27< fabi> Would you say that the Forum is a part of communication? In fact the most important Wesnoth has? Are you just a living sock puppet?
20160224 11:55:40< vultraz> I said, we're not having this argument
20160224 11:57:34< vultraz> But for the record, I am not a living sock puppet 
20160224 11:58:05< fabi> :-) Good to know. You should wear a t-shirt that says it loud.
20160224 12:00:46< Elvish_Hunter> zookeeper: according to the Wayback Machine, it was deleted between 16 Jan 2016 and 11 Feb 2016.
20160224 12:01:22< iceiceice> if there is some sort of general campaign to delete forum regions, it would be kind of nice if we could have a schedule maybe
20160224 12:01:41< iceiceice> so we can take steps to back them up / savor them before they are gone perhaps
20160224 12:02:28< iceiceice> had it happened historically the forum regions are just deleted without notice every few months?
20160224 12:02:39< iceiceice> if it did i suppose i didn't notice it :O
20160224 12:02:40< fabi> no
20160224 12:02:49< fabi> we never deleted whole regions
20160224 12:02:51< zookeeper> shadowm doesn't tend to go rogue often enough to warrant assuming that he just went and deleted it without saying anything to anyone, i'd think.
20160224 12:03:22< fabi> what a crap
20160224 12:03:27< iceiceice> i mean it could be mentioned in "website"
20160224 12:03:35< fabi> shadowm is the rogue in person
20160224 12:03:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160224 12:04:18< iceiceice> hmmmm
20160224 12:04:36< vultraz> he might have mentioned something to me at one point
20160224 12:04:38< iceiceice> on jan 24 the changelog entry is "
20160224 12:04:38< iceiceice> Re: Forums Changelog
20160224 12:04:38< iceiceice> Unread postby shadowm » January 24th, 2016, 3:34 pm
20160224 12:04:38< iceiceice> The Forum Games forum has been restored.
20160224 12:04:38< iceiceice> Also, a lot of structural changes I may write down later maybe eventually I don’t know."
20160224 12:04:57< vultraz> also, there are backups 
20160224 12:05:00< iceiceice> i guess "Experimental" going away is one of those
20160224 12:05:47< fabi> oh
20160224 12:05:58< fabi> I guess that explains everything.
20160224 12:06:36< zookeeper> oh, right, forums changelog. forgot about that.
20160224 12:06:45< zookeeper> i guess we'll find out next week.
20160224 12:09:02-!- travis-ci [~travis-ci@ec2-54-145-78-22.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 12:09:03< travis-ci> wesnoth/wesnoth#8596 (master - b76813c : Charles Dang): The build passed.
20160224 12:09:03< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111451021
20160224 12:09:03-!- travis-ci [~travis-ci@ec2-54-145-78-22.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 12:09:44-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has left #wesnoth-dev ["Ciao!"]
20160224 12:09:56< fabi> vultraz: please install a way to easily clone the forum, so it can be forked without much fuzz.
20160224 12:10:31< iceiceice> gnight folks
20160224 12:10:42< fabi> iceiceice: bye :-)
20160224 12:13:07-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20160224 12:13:38< vultraz> A. why in god's name would I do that
20160224 12:13:48< vultraz> B. even if I knew how, which I don't, I wouldn't
20160224 12:14:06< vultraz> C. if I knew how, I'd upgrade the forums to something other than phpbb
20160224 12:14:24< fabi> vultraz: just dump the database
20160224 12:16:07< fabi> A) I think forking a projects' resources is just as legitimate as forking GPL code. So it pretty much would be in the spirit of open source to do so. 
20160224 12:20:46< vultraz> I don't care about the spirit of open source 
20160224 12:21:02< fabi> Not surprised to hear that.
20160224 12:21:09< vultraz> and it would not be "legitimate" to release all of the forum database's contents to the world
20160224 12:22:14< zookeeper> wow, there's a lot of missing unrenamable=yes's in some campaigns...
20160224 12:22:35< zookeeper> vultraz, maybe you'll be delighted to hear that i'll be adding some
20160224 12:26:24< fabi> vultraz: So what is it worth that there is a backup of the missing forum if you and your henchmen don't allow me to have it?
20160224 12:27:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 12:28:15< Ravana_> fabi: https://forums.wesnoth.org/viewforum.php?f=36
20160224 12:28:43< fabi> Ravana_:  Thank you :-)
20160224 12:31:53< vultraz> fabi: so the administrators can restore the forums should something happen?
20160224 12:31:56< vultraz> why else?
20160224 12:32:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20160224 12:32:48< irker475> wesnoth: ln-zookeeper wesnoth:master b338a894b704 / data/campaigns/ (14 files in 6 dirs): Added some missing unrenamable=yes keys and removed unnecessary ones https://github.com/wesnoth/wesnoth/commit/b338a894b704e7eb55a7c6a59eee3c885f57a106
20160224 12:32:59< vultraz> zookeeper: just a note: if we adhere by your decision to allow custom naming of most units, we cannot allow SUF filtering by name
20160224 12:33:24< vultraz> can't have both
20160224 12:33:38< zookeeper> of course we can, you just have to be aware of that when deciding what units to filter on.
20160224 12:34:32-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160224 12:34:48< zookeeper> that's like saying that if we allow player's units to die, we can't allow messages because the player might have gotten the speaker killed already
20160224 12:35:09< fabi> Ravana_: The forum is not linked to from toplevel anymore but it is still open to read and write?
20160224 12:35:23< Ravana_> it is liked from top->ideas->
20160224 12:35:27< zookeeper> it's not exactly rocket science to be aware of the possibility when writing your filters :P
20160224 12:35:34< Ravana_> just like some mainline feedback
20160224 12:36:52< fabi> I see :-)
20160224 12:40:04< zookeeper> vultraz, looks like everyone in at least SoF and SotBE are renamable currently :p
20160224 12:40:21< zookeeper> also in tutorial :>
20160224 12:42:34< vultraz> change that immediately!
20160224 12:47:09< zookeeper> cool, you could also rename Konrad II near EI's end
20160224 12:49:03< zookeeper> but! i won't change that, because he's never referred to by name, only as "your highness" and so on :> the epilogue could be hilarious with him having a stupid name
20160224 12:49:17< zookeeper> that almost qualifies as an easter egg
20160224 12:49:25-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 252 seconds]
20160224 12:49:40< zookeeper> did i give him a stroke?
20160224 12:53:22-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160224 12:54:45-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160224 13:02:23< irker475> wesnoth: ln-zookeeper wesnoth:master a33e37a2fc47 / data/campaigns/ (11 files in 4 dirs): Added unrenamable=yes to some more units https://github.com/wesnoth/wesnoth/commit/a33e37a2fc47b3086dbb1734f038de2e815ae5f9
20160224 13:02:55< zookeeper> in NR, only tallin has it
20160224 13:04:55-!- travis-ci [~travis-ci@ec2-54-145-78-22.compute-1.amazonaws.com] has joined #wesnoth-dev
20160224 13:04:56< travis-ci> wesnoth/wesnoth#8597 (master - b338a89 : ln-zookeeper): The build has errored.
20160224 13:04:56< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/111464904
20160224 13:04:56-!- travis-ci [~travis-ci@ec2-54-145-78-22.compute-1.amazonaws.com] has left #wesnoth-dev []
20160224 13:21:42-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 13:26:13-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20160224 13:50:13< celmin|sleep> The only problem with Boost.Spirit is that the documentation is so terrible that it forced me into using Spirit.Classic instead. And I realize Elvish_Hunter isn't here at the moment, but I think a wmllint rule for name= in filters would be great, if you can manage it (don't forget that filters appear in a plethora of different tags, so it might be quite difficult).
20160224 13:57:50-!- celmin|sleep is now known as celticminstrel
20160224 14:07:16-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160224 14:08:36-!- gfgtdf [~chatzilla@f054061068.adsl.alicedsl.de] has joined #wesnoth-dev
20160224 14:09:20-!- gfgtdf [~chatzilla@f054061068.adsl.alicedsl.de] has quit [Client Quit]
20160224 14:15:49-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 14:16:26< mattsc> celticminstrel: what’s wrong with AiWML?
20160224 14:17:47< celticminstrel> Uh, what was it...
20160224 14:18:36< mattsc> You said it looks like I had not worked on it yet, whereas I thought it was pretty close (which is not the same as done, of course)
20160224 14:18:55< celticminstrel> I... don't remember.
20160224 14:19:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160224 14:19:31< celticminstrel> I guess the last two sections are kind of silly, since they're so short, but...
20160224 14:20:15< mattsc> Yes — that bothers me a little too, but their content is better on other pages while it still makes sense to at least link to them from here.
20160224 14:20:33< mattsc> If you have a suggestion for how to do that better, I wouldn’t mind.
20160224 14:20:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160224 14:20:51< mattsc> Oh, wow, I have a user wiki page ...
20160224 14:21:10< celticminstrel> Hmm, the first section says "See the bottom of this page". Would it make sense to just move that content up to there?
20160224 14:21:13< mattsc> Not updated since 2011, no wonder I didn’t remember
20160224 14:21:36< mattsc> yeah, possible
20160224 14:21:41< mattsc> *bly
20160224 14:22:09< mattsc> Either way, if it’s a relatively minor thing like that, I’m not worried.
20160224 14:22:25< mattsc> At the moment I m still working on much more major revisions on other pages.
20160224 14:22:40< celticminstrel> I can't remember what it was that made me say it before, but I can't see anything terribly wrong right now, so whatever.
20160224 14:23:32< mattsc> And I have checked out the Practical Guide page now.  I agree that Create should link to Wesnoth_AI and the rest of the links can be ignored.
20160224 14:24:05< mattsc> So if you want to do that and move it to the category vultraz suggested for deletion, I wouldn’t mind.  :)
20160224 14:25:54< mattsc> And I’ll go back to LuaAI, I guess …
20160224 14:27:54-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160224 14:32:43< celticminstrel> I think I'd recommend leaving the Chinese version alone for a bit, but the English version is basically deleted already and just needs to be finalized.
20160224 14:36:15< mattsc> Agreed; my thoughts as well.
20160224 14:41:54-!- aidanhs [~aidanhs@81.4.110.234] has quit [Ping timeout: 276 seconds]
20160224 14:45:56-!- aidanhs [~aidanhs@81.4.110.234] has joined #wesnoth-dev
20160224 14:53:40-!- louis94 [~~louis94@91.178.241.231] has joined #wesnoth-dev
20160224 14:53:52-!- louis94 [~~louis94@91.178.241.231] has quit [Client Quit]
20160224 14:55:38-!- boucman_work [~jrosen@80-15.80-90.static-ip.oleane.fr] has joined #wesnoth-dev
20160224 14:55:38-!- boucman_work [~jrosen@80-15.80-90.static-ip.oleane.fr] has quit [Changing host]
20160224 14:55:38-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160224 14:56:15-!- louis94 [~~louis94@91.178.241.231] has joined #wesnoth-dev
20160224 14:57:07-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 248 seconds]
20160224 15:04:57-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160224 15:09:28< mattsc> celticminstrel: Question: I am not entirely sure what best to do with AIModule.
20160224 15:09:36-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160224 15:09:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 15:10:16< mattsc> Most of its content is explained elsewhere and in a much more understandable way, in my opinion.  However, it does provide a more developer oriented way of presenting the information.
20160224 15:10:39< celticminstrel> Developer meaning eg me?
20160224 15:10:43< mattsc> On the other hand, it is probably outdated and I really don’t want to spend the time updating all that stuff that probably nobody will ever look at.
20160224 15:10:54< mattsc> Meaning anybody who wants to meddle with the C++ code.
20160224 15:11:09< mattsc> Any opinion on what to do with it?
20160224 15:11:38< mattsc> The options, in my opinion, are either keep it as is, with a warning that it might not quite be up to date, or delete it.
20160224 15:13:19-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit [Quit: Bye]
20160224 15:13:38-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev
20160224 15:14:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20160224 15:16:22< celticminstrel> So... section 1 seems kinda useless. Section 2 might be useful; is that information anywhere else? Section 3 I assume you've integrated into the other pages.
20160224 15:16:45< mattsc> Also, I am not going to clean up AI Arena at this time, because I think there are better ways to do the testing.
20160224 15:16:51< celticminstrel> And section 4 is basically empty.
20160224 15:17:00< mattsc> I might do so after I clean up data/ai/
20160224 15:17:39< celticminstrel> Where has the information from section 4 been moved to?
20160224 15:17:44< mattsc> Agreed on Sections 1, 3 and 4.
20160224 15:17:49< mattsc> And no, on Section 2
20160224 15:18:07< mattsc> It has not been moved.  It’s always been like that.
20160224 15:18:36< celticminstrel> You said it was explained elsewhere.
20160224 15:18:48< celticminstrel> Oh, sorry, section 3.
20160224 15:18:50< celticminstrel> Not 4.
20160224 15:19:05< mattsc> Oh …
20160224 15:19:37< mattsc> Most of it has gone here: https://wiki.wesnoth.org/Wesnoth_AI_Framework
20160224 15:20:16< celticminstrel> I can see that page doesn't cover the [engine] syntax.
20160224 15:21:02-!- prkc [~prkc@gateway/vpn/privateinternetaccess/prkc] has joined #wesnoth-dev
20160224 15:22:33< mattsc> True.  That’s because you don’t need engines any more.  And if you want to use them (only needed in Lua), they are described in the old Lua Howto.
20160224 15:22:59-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20160224 15:23:12< celticminstrel> The only reason I can think of that you'd need to use them is if you want to set up some initialization type stuff in the Lua engine (either code or [data]).
20160224 15:24:15< mattsc> This good enough?  https://wiki.wesnoth.org/Lua_AI_Legacy_Methods_Howto
20160224 15:24:23< mattsc> search for [engine] on that page
20160224 15:24:43< celticminstrel> I dunno. Maybe?
20160224 15:24:56< mattsc> While I don’t recommend using that method any more, I don’t intend to delete it.
20160224 15:25:33< celticminstrel> What happens to the value returned from the [engine] code?
20160224 15:25:53< mattsc> It’s available as an upvalue to the CAs
20160224 15:26:06< mattsc> the old-style CAs, not the new ones
20160224 15:26:25< celticminstrel> Is it passed to them as an argument, or what?
20160224 15:27:05< mattsc> Check out the bottom of the code box in the first section on that howto page I just linked to
20160224 15:28:15-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20160224 15:29:08< celticminstrel> I guess the (...) is referring to that value?
20160224 15:29:30< celticminstrel> It's not very obvious.
20160224 15:29:38< mattsc> yes, that’s what crab, nephro et al. called the upvalue.
20160224 15:29:52< mattsc> I don’t know if that’s the correct term, I am just echoing them.
20160224 15:29:56< celticminstrel> Ah.
20160224 15:30:01< mattsc> Also LuaWML says something about that, IIRC.
20160224 15:30:09< celticminstrel> I don't know if it's correct, but I think of it as arguments to a function.
20160224 15:30:15< celticminstrel> Okay, so...
20160224 15:30:25< celticminstrel> This thing about old-style vs new-style Lua CAs.
20160224 15:30:36-!- ancestral [~ancestral@63.92.240.233] has quit [Client Quit]
20160224 15:30:50< celticminstrel> Old-style uses evaluation= and execution= keys and receives as upvalue whatever the engine code returned, right?
20160224 15:31:11< mattsc> yes
20160224 15:31:39< celticminstrel> While new-style uses a location key and receives as upvalue a) the AI functions table, b) the config defined by exec_parms or eval_parms, and c) the self variable containing the data table.
20160224 15:31:56< mattsc> correct
20160224 15:32:36< celticminstrel> Hmm. I wonder what would happen if you defined the [engine] code to return something, but then used new-style CAs... would it receive that instead of the AI functions table as the first parameter...
20160224 15:32:46< mattsc> I’m going to start a second conversation, I also don’t really know what to do with the remainder of https://wiki.wesnoth.org/Customizing_AI_in_Wesnoth_1.8
20160224 15:33:08< mattsc> I don’t think so.  I haven’t seen a point in trying.
20160224 15:33:42< mattsc> As far as I am concerned, the message I would like to give to folks is: here’s the method you should use (which is the new-style CAs)
20160224 15:33:53< celticminstrel> Okay, so, from the overview on AI Module, everything except the [engine] syntax is covered in Wesnoth AI Framework.
20160224 15:34:05< celticminstrel> The list of available engines isn't particularly important.
20160224 15:34:15< mattsc> You don’t even need to know that there are other methods.  But if you come across code written using the old method, we still provide documentation as to what that means.
20160224 15:34:56< mattsc> yes
20160224 15:34:58< celticminstrel> The list of available aspects probably isn't really important either, since it's just "composite_aspect" and "simple_aspect". It's pretty much covered by Wesnoth AI Framework, anyway.
20160224 15:35:43< celticminstrel> Really, "composite_aspect" is just [aspect] while "simple_aspect" is [facet]. At least, as far as the end-user is concerned.
20160224 15:36:17< celticminstrel> Oh, there's no mention of [default] facets in Wesnoth AI Framework, though.
20160224 15:36:26< celticminstrel> Oh wait.
20160224 15:36:42< celticminstrel> It's mentioned in a listing, but not explained.
20160224 15:38:02< mattsc> I can make that clearer (add it to the list of attricutes) when I do my final read-through of that page
20160224 15:38:28< celticminstrel> The list of available stages could maybe be mentioned in a sentence somewhere, though the FAI ones are probably not worth it, and nor is recruitment.
20160224 15:39:04< mattsc> https://wiki.wesnoth.org/Wesnoth_AI_Framework#Legacy_Stages
20160224 15:39:08< celticminstrel> Lua and Fallback stages could be mentioned, something like "Other possible stages are the lua stage with a code key and the fallback stage which contains an [ai] tag".
20160224 15:39:12< celticminstrel> Ah., I missed that.
20160224 15:39:30< celticminstrel> Okay, yeah, that's more than I was expecting.
20160224 15:39:33< mattsc> and recruitment stage is mentioned right above that
20160224 15:40:15< celticminstrel> I don't think it's necessary for a fallback stage to fall back to older AI.
20160224 15:40:16-!- louis94 [~~louis94@91.178.241.231] has quit [Quit: Konversation terminated!]
20160224 15:40:46< celticminstrel> I'm not sure that fallback is necessary, but only because I don't really see the difference between installing a stage in a fallback and just installing that stage directly.
20160224 15:40:58< celticminstrel> Anyway...
20160224 15:41:08< mattsc> yes, exactly
20160224 15:41:18< celticminstrel> The list of candidate actions should definitely be somewhere. I think I saw it in RCA AI, so that should be fine probably.
20160224 15:41:37< mattsc> yes, it is
20160224 15:41:55< mattsc> and it includes a linnk to how to find those horrendously complicated names
20160224 15:41:57< celticminstrel> The only thing is that the symbolic names do need to be mentioned somewhere, since they're what you use in [modify_ai].
20160224 15:42:06< celticminstrel> Ah, well, fair enough I guess?
20160224 15:42:18< mattsc> That’s my take on it.
20160224 15:42:34< celticminstrel> Where is this link?
20160224 15:43:58< mattsc> https://wiki.wesnoth.org/RCA_AI#Available_Candidate_Actions
20160224 15:44:23< celticminstrel> Ah, okay.
20160224 15:44:42< mattsc> Coming back to Customizing in 1.8, everything but the very last section is covered somewhere else by now.
20160224 15:44:53< celticminstrel> Actually, it's the ID that's normally used in [modify_ai], right?
20160224 15:44:57< mattsc> Not necessarily using exactly the same examples, but who cares.
20160224 15:45:10< mattsc> Any opinion on the last section?  Is that worth keeping somewhere?
20160224 15:45:21< celticminstrel> But I guess you'd need to use the name if you wanted to add the aspect for some reason, with a different score or something.
20160224 15:45:36< mattsc> The id is used in [modify_ai], but the name identifies to the engine which CA this is.
20160224 15:45:53< mattsc> yes
20160224 15:45:57< celticminstrel> Wait, I somehow got confused. We're not even talking about aspects here.
20160224 15:46:16< mattsc> No; aspects don’t have scores.  I figured you had just mistyped.  :)
20160224 15:46:33< celticminstrel> {AI_NEW_SYNTAX} will shortly be a no-op.
20160224 15:47:15< mattsc> There’s the same type of link to available aspects at the bottom of the page too.
20160224 15:47:58< mattsc> I need to head to work shortly.  Did you see my question on the Customizing in 1.8 page?
20160224 15:48:04< mattsc> Or did it get lost in all the other stuff?
20160224 15:48:28< celticminstrel> Yeah, reading the section now.
20160224 15:48:53< mattsc> Other than than, I think I am down to Creating_Custom_AIs and Wesnoth_AI for pages which need significant work.
20160224 15:49:08< mattsc> And all pages need a final clean-up, of course.
20160224 15:50:46< celticminstrel> That last section is pretty badly written... there might be a bit of useful stuff in there for devs, but probably not much.
20160224 15:51:13< mattsc> I think it’s mostly a copy-and-paste from irc.
20160224 15:51:22< mattsc> It was there already the first time I looked at that page.
20160224 15:51:33< mattsc> So, copy it somewhere?  Or just get rid of it?
20160224 15:52:01< mattsc> Besides that, that page can be tossed too (after checking what links there)
20160224 15:52:52< mattsc> Ah, I did that already (redirecting the links; there were a lot originally)
20160224 15:53:40< celticminstrel> I haven't looked closely at the source yet, but I suspect everything it says could be learned by going through the source...
20160224 15:55:26< mattsc> Yes, it can; I’d say, lets just get rid of it then
20160224 15:55:47< mattsc> Also, what’s your final verdict on AI Module?  Delete also?
20160224 15:56:48< celticminstrel> The only remotely useful thing there was the list of what each source file is for, but that might not be significantly more useful than looking at the files themselves.
20160224 15:57:01< celticminstrel> Also the [engine] syntax, but I guess that's covered elsewhere too.
20160224 15:57:10< celticminstrel> On some Lua legacy page or something.
20160224 15:57:20< mattsc> yep
20160224 15:57:35< mattsc> okay — I have 8 pages marked for deletion then
20160224 15:58:09< mattsc> I’ll go through the links and add them to the deletion category in a little while
20160224 15:58:35< mattsc> celticminstrel: thanks for all the help and discussion
20160224 15:59:38< mattsc> I’ll be back on in about half an hour
20160224 16:00:06-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20160224 16:01:40-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160224 16:02:07-!- boucman_work [~jrosen@80-15.80-90.static-ip.oleane.fr] has joined #wesnoth-dev
20160224 16:02:07-!- boucman_work [~jrosen@80-15.80-90.static-ip.oleane.fr] has quit [Changing host]
20160224 16:02:07-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160224 16:02:32-!- irker475 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160224 16:04:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 16:08:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160224 16:11:44-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160224 16:13:52-!- Kwandulin [~Miranda@p200300760F0BC5832CCC877D7A01ED7D.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160224 16:20:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 16:24:33-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20160224 16:25:06-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 244 seconds]
20160224 16:30:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 16:33:07-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160224 16:43:08-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160224 16:53:32-!- gfgtdf [~chatzilla@f054061068.adsl.alicedsl.de] has joined #wesnoth-dev
20160224 16:55:04< gfgtdf> hmm i updated boost but it sisnt solve my problem (which is that the ordering iof translateable strings is wrong, i want the 'A' < 'a' < 'Ä' < 'ä'< 'B' < 'b' < 'C' .. order)
20160224 16:56:15< celticminstrel> I would expect AÄaä...
20160224 16:56:30< celticminstrel> ie, collation with accents ignored.
20160224 16:56:53< celticminstrel> Though obviously in some languages, that's wrong - Spanish ñ for example.
20160224 16:56:56< gfgtdf> celticminstrel: hmm y thats fine too, im not realyl an expert
20160224 16:57:28< gfgtdf> celticminstrel: oh it seems leik i deactivated collate functions in gettext_boost.cpp manually becsue i thought thats not needed
20160224 16:58:00< gfgtdf> celticminstrel:mostlikeley i didnt even need to rebuild boost hmm.
20160224 16:58:34-!- louis94 [~~louis94@91.178.241.231] has joined #wesnoth-dev
20160224 16:59:58< gfgtdf> celticminstrel: you know what "std::collate supports only one level of collation" means ?
20160224 17:00:41< celticminstrel> No idea.
20160224 17:03:37-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has quit [Quit: Leaving]
20160224 17:04:04-!- louis94 [~~louis94@91.178.241.231] has quit [Quit: Konversation terminated!]
20160224 17:07:34-!- irker717 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160224 17:07:34< irker717> wesnoth: gfgtdf wesnoth:master b282ef7236ce / src/ (6 files in 3 dirs): use game_data* instead of boost::optional<game_data&> in tod_manager https://github.com/wesnoth/wesnoth/commit/b282ef7236cee438b3e6dde650a4c7f82bfda485
20160224 17:07:34< irker717> wesnoth: gfgtdf wesnoth:master bcb06e4a7a74 / src/ (gettext.hpp gettext_boost.cpp lua/lvm.cpp scripting/lua_common.cpp): fix lua string and tstring comparision https://github.com/wesnoth/wesnoth/commit/bcb06e4a7a74b894b5247ad867df905722d334a4
20160224 17:09:40< celticminstrel> gfgtdf: You could've simply included game_data.hpp instead.
20160224 17:09:46< gfgtdf> celticminstrel: i wonder whether i shoudl put icu libaray to the list of optional dependencies
20160224 17:09:59< celticminstrel> But, I guess since it was an optional reference, maybe it's not too bad that way.
20160224 17:10:07< gfgtdf> celticminstrel: well yes i could, but i dont realyl see advantages of that over my solution
20160224 17:12:19< celticminstrel> Lua strings can contain nulls?
20160224 17:12:37-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160224 17:13:26< celticminstrel> Does Lua automatically implement >= <= and != based onf < > and ==?
20160224 17:13:35< celticminstrel> ^-f
20160224 17:13:46< celticminstrel> Anyway, looks good to me.
20160224 17:14:48< celticminstrel> Oh, wait, I guess it would be > >= and != based on < <= and ==
20160224 17:15:17< celticminstrel> I think I'd prefer testing <= 0 rather than < 1, even though it's the same, but not important.
20160224 17:15:29< celticminstrel> Not important enough to change it now that it's been pushed.
20160224 17:21:20< Yaiyan> Should rhoders open a pull-request with his new animations, or?
20160224 17:28:14-!- TC01 [~quassel@london.acm.jhu.edu] has joined #wesnoth-dev
20160224 17:28:59-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20160224 17:31:00< Ravana_> Lua has <= and >= but != is called ~=
20160224 17:32:33< elias> ~= reminds me of Perl
20160224 17:39:11< celticminstrel> Right, right. Question still stands though.
20160224 17:39:22< celticminstrel> elias: If I remember right, isn't it =~ in Perl?
20160224 17:39:30< celticminstrel> (And completely different meaning.)
20160224 17:44:38-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20160224 17:48:17-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 244 seconds]
20160224 17:59:31-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160224 18:10:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 18:17:47< gfgtdf> vultraz: how likeley do you think it to have the mp screen in gui2 in 1.14 ?
20160224 18:21:54-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 18:22:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 18:22:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 18:41:14-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160224 18:57:30< mattsc> Note to self: I need to make sure that we distinguish between goals and targets on the wiki pages
20160224 18:57:48< mattsc> Those are almost interchangeable, but not quite.
20160224 18:57:56< irker717> wesnoth: gfgtdf wesnoth:master bcf2710cc64e / src/game_initialization/multiplayer_ui.cpp: add a comment https://github.com/wesnoth/wesnoth/commit/bcf2710cc64e586f6f865bb8ebaed680417ae512
20160224 18:57:58< irker717> wesnoth: gfgtdf wesnoth:master ac6439efdc26 / src/hotkey/command_executor.cpp: fix lua console showing twice https://github.com/wesnoth/wesnoth/commit/ac6439efdc26cf5be0a9341e25fa9aa44f69543d
20160224 19:00:04-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160224 19:02:06-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160224 19:02:09-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160224 19:02:20-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20160224 19:05:36< vultraz> gfgtdf: if we fix the label size bug, quite possible.
20160224 19:11:59< celticminstrel> mattsc: Wait what?
20160224 19:17:28< mattsc> celticminstrel: technically speaking, goals are the concepts of what to head toward; whereas targets are the actual hex coordinates of those.
20160224 19:18:25< mattsc> More loosely, the term target is also used for goals, but goals should not be used for that technical target definition.
20160224 19:21:10-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 19:24:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 19:35:24-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20160224 19:38:41-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160224 19:38:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 19:38:46-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has joined #wesnoth-dev
20160224 19:48:16-!- ancestral [~ancestral@209.181.254.220] has joined #wesnoth-dev
20160224 19:48:58< mattsc> celticminstrel: I decided to add the [engine] syntax to the AI Framework page after all, but I am confused about something.
20160224 19:49:18< mattsc> All the examples I see defining a Lua engine use name=lua
20160224 19:49:33< mattsc> However, this  https://github.com/wesnoth/wesnoth/blob/master/src/ai/contexts.cpp#L662
20160224 19:49:43 * vultraz pings celticminstrel
20160224 19:49:54 * celticminstrel pongs vultraz 
20160224 19:50:05< mattsc> looks like the engine name is actually defined by [engine]engine=
20160224 19:50:10< mattsc> what am I missing?
20160224 19:50:35< vultraz> I'm finally getting back to the widget groups thing
20160224 19:52:47< vultraz> celticminstrel: is this the correct way to use a class template http://pastebin.com/6V9M3eLw
20160224 19:53:13< vultraz> (fixed the highlighting)
20160224 19:54:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 19:56:38< ancestral> TIL there’s a Konrad imposter
20160224 19:56:54-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 19:57:01-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 19:58:34< zookeeper> ancestral, better late than never?
20160224 19:59:15< ancestral> (Not that Konrad is an imposter, but that there’s an imposter *of* Konrad)
20160224 19:59:27< zookeeper> oh.
20160224 19:59:35< ancestral> :-P
20160224 19:59:52< zookeeper> where's that?
20160224 20:00:56< vultraz> hm
20160224 20:01:00< vultraz> actually
20160224 20:01:06< vultraz> why am I using a class template again
20160224 20:01:28< vultraz> celticminstrel: please advise 
20160224 20:01:38< ancestral> zookeeper: data/ai/micro_ais/scenarios/patrols.cfg
20160224 20:02:31< zookeeper> ah.
20160224 20:03:18< mattsc> Somebody apparently had a strange sense of humor!
20160224 20:04:06< zookeeper> scandalous
20160224 20:06:05< ancestral> Wesnoth was built on a lie!
20160224 20:06:22< ancestral> vultraz: I have a question you might be able to guide me on
20160224 20:07:12< SigurdFD> zookeeper: I agree with the renaming reasoning. Though, would it appropriate to make the student unrenameable in Tutotrial for the sake of continuty with HttT?
20160224 20:07:58< vultraz> zookeeper, SigurdFD: yes, please do that
20160224 20:08:01< celticminstrel> mattsc: Uh, isn't that line asking for the engine attribute?
20160224 20:08:16< vultraz> they're referred to by name in some lines
20160224 20:08:29< ancestral> vultraz: I’m looking for the script or code that handles [unit_profile]. Would it maybe be hanign out in data/gui/default/window ?
20160224 20:08:42< celticminstrel> ...oh wait, is that the code that loads an [engine]?
20160224 20:08:49< ancestral> *hanging
20160224 20:09:36< mattsc> that’s what I think it is; and that line reads the engine attribute into engine_name.  I think.  Oh …
20160224 20:09:40< vultraz> ancestral: unit profile?
20160224 20:09:45< SigurdFD> looked closer, there's also dialog that referes to Konrad & Li'sar by name, so that'd be yes.
20160224 20:09:56< celticminstrel> I think it looks like it's fetching the engine required to interpret a given stage or aspect.
20160224 20:10:09< SigurdFD> zookeeper: ^
20160224 20:10:23< mattsc> Hmm, you process a Lua [engine] tag using the cpp engine.  I think.
20160224 20:10:24< ancestral> Yeah, it’s used in Themes, but maybe elsewhere too? (Help?)_
20160224 20:10:37< SigurdFD> vultraz: ok, working on a pr now.
20160224 20:10:38< ancestral> vultraz: Maybe you don’t know
20160224 20:10:41< vultraz> ancestral: I have absolutely no idea
20160224 20:10:51< vultraz> ancestral: possibly somewhere in src/reports or something
20160224 20:10:53< ancestral> Okay, no problem
20160224 20:10:57< ancestral> I did see reports
20160224 20:10:59< vultraz> ancestral: but it's certainly not gui2 if its in themes
20160224 20:10:59< ancestral> Actually
20160224 20:11:02< celticminstrel> vultraz: I think you were using a class template because you wanted the groups to be generalizable to widgets other than toggle buttons.
20160224 20:11:07< ancestral> What is reports.cpp?
20160224 20:11:20< celticminstrel> I'd suggest not giving T a default value.
20160224 20:11:23< vultraz> don't ask me that either :P
20160224 20:11:27< mattsc> celticminstrel: yeah, maybe.  I’m just trying to make sense of the {engine] tag description on AI Module.
20160224 20:11:28< zookeeper> SigurdFD, yes, i just didn't get around to that yet
20160224 20:11:35< vultraz> celticminstrel: hm... or I could just use tselectable_ like i was doing before
20160224 20:11:41< vultraz> or twidget
20160224 20:11:44< celticminstrel> Isn't reports.cpp the file that writes the contents of the sidebar?
20160224 20:12:33< celticminstrel> vultraz: I'd say you need to figure out exactly how you want the groups to work.
20160224 20:13:20< celticminstrel> If you use tselectable_, then it'll be limited to classes that implement that interface. If you use twidget, it'll allow any type of widget. In both of the latter cases, it could potentially contain multiple types of widget.
20160224 20:13:31-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160224 20:15:29< celticminstrel> I think a homogenous group makes more sense, though that's impossible to completely enforce when using pointers (and not using pointers is probably difficult).
20160224 20:15:51< vultraz> yes, they make more sense
20160224 20:17:39< celticminstrel> So, in pseudo-code, how would you like the groups to be used when defining a dialog?
20160224 20:18:46< celticminstrel> BTW, since you're making it a template, you probably won't need a source file.
20160224 20:19:02< celticminstrel> Templates are normally header-only.
20160224 20:20:54< celticminstrel> It's sort of possible to split it off into a source file, but generally with some kind of weird include structure.
20160224 20:21:28< celticminstrel> (Either you include the source file instead of the header, or the header includes the source file.)
20160224 20:21:52< celticminstrel> (Also, they normally don't have .cpp extensions - something like .tpp or .tcc is more common.)
20160224 20:22:00< ancestral> celticminstrel: That would make a ton of sense (the sidebar)
20160224 20:22:20 * celticminstrel poke vultraz 
20160224 20:22:21< ancestral> (I’m trying to fix my own bug I filed)
20160224 20:23:32< SigurdFD> zookeeper, vultraz: any issue with me adding another commit to the PR to change Delfador's traits in T to match HttT?
20160224 20:25:40< ancestral> Hmm, I wonder if it doesn’t scale up, and instead only scales down…
20160224 20:25:40-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160224 20:25:41-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160224 20:27:09-!- ancestral [~ancestral@209.181.254.220] has quit [Quit: i go nstuf kthxbai]
20160224 20:27:33< zookeeper> SigurdFD, eh, let me check...
20160224 20:30:08< zookeeper> SigurdFD, i guess he can be intelligent in the tutorial.
20160224 20:30:44< vultraz> celticminstrel: I'm thinking of just keeping it selectable_
20160224 20:31:00< celticminstrel> vultraz: So what about my question?
20160224 20:31:02< SigurdFD> it's odd, it seems he always ends up with itelligent, resilent right now in T. Was going to change it to loyal, itelligent
20160224 20:31:10< vultraz> ok
20160224 20:31:12< vultraz> well
20160224 20:31:25< vultraz> in preshow, you'd add the widget to a group
20160224 20:32:01< zookeeper> SigurdFD, i don't think he should be loyal though, as he's not even on your side.
20160224 20:32:23< vultraz> and the dialog would handle any special usecases of the groups (since I'm implementing iterator pairs, now)
20160224 20:32:37< vultraz> but widgets could implement their own routines
20160224 20:32:54< SigurdFD> ok, I'll set it to just intelligent then
20160224 20:33:14< vultraz> for example
20160224 20:33:30< vultraz> I'm thinking tselectable should implement a 'get active widget' function
20160224 20:33:34< vultraz> not tgroup
20160224 20:33:48< vultraz> tgroup should just be the generic class
20160224 20:36:22< zookeeper> SigurdFD, and yeah i don
20160224 20:36:37< zookeeper> 't even know how it's possible that it's not been noticed, but he doesn't have random_traits=no so he ends up with random traits -.-
20160224 20:38:05< SigurdFD> is it random for you? I always get intellegent, resilent before making my PR, does that mean there's a seed error with the RNG?
20160224 20:38:37< vultraz> celticminstrel: but since I want tgroup to be generic, maybe it should be twidget-based..
20160224 20:39:28< celticminstrel> Or templated.
20160224 20:39:34< zookeeper> SigurdFD, yeah i got intelligent,resilient three times in a row
20160224 20:39:37< vultraz> blah
20160224 20:39:43< SigurdFD> 6 on my end
20160224 20:39:50< zookeeper> SigurdFD, i'm... not sure why that is
20160224 20:39:53< vultraz> celticminstrel: ok so I think I'm a little confused as to class templates
20160224 20:40:03< vultraz> function templates are basically dynamic types, right
20160224 20:40:15< celticminstrel> Uhhh. Function templates are functions, not types.
20160224 20:40:21< vultraz> er
20160224 20:40:23< vultraz> right
20160224 20:40:25< vultraz> yes
20160224 20:40:27< celticminstrel> Also, there's nothing whatsoever dynamic about templates.
20160224 20:40:27< vultraz> yes..
20160224 20:40:30< celticminstrel> They're entirely static.
20160224 20:40:45< vultraz> I meant generic, not dynamic
20160224 20:40:47< vultraz> sorry
20160224 20:41:15< celticminstrel> Templates are generic, yes. That's pretty much what they do.
20160224 20:41:40< vultraz> like, I understand template <typename T> foo(T bar). But I'm not sure I understand the templates in a class context
20160224 20:41:56< celticminstrel> It's more or less the same.
20160224 20:42:03< zookeeper> SigurdFD, even if i place another generic (human) unit before delfador, they get intelligent,resilient, and then delf gets something else.
20160224 20:42:21< celticminstrel> Except you're defining a class instead of a function.
20160224 20:42:39< celticminstrel> So, you'd construct the group during pre_show?
20160224 20:43:03< vultraz> so if I have template<class T> class foo { public : }, what does T do
20160224 20:43:03< celticminstrel> What does constructing the group entail?
20160224 20:43:48< celticminstrel> Nothing. It's unused.
20160224 20:44:01< celticminstrel> But that class is pointless anyway, since it's empty.
20160224 20:44:06< zookeeper> SigurdFD, certainly an interesting detail to look into.
20160224 20:44:34< vultraz> yes, but I mean what could T do
20160224 20:44:39< vultraz> what does the T allow you to do
20160224 20:45:08< celticminstrel> It can do anything you can do with a class. You can inherit from it, declare variables and functions with it, etc.
20160224 20:45:45< celticminstrel> If I recall correctly, isn't there something like connect_signal_notify_modified?
20160224 20:46:00< vultraz> so you could say class foo, public T { } ?
20160224 20:46:02-!- louis94 [~~louis94@91.178.241.231] has joined #wesnoth-dev
20160224 20:46:13< vultraz> celticminstrel: yes
20160224 20:46:27< celticminstrel> You could indeed say "class foo : public T {};"
20160224 20:46:36< celticminstrel> Probably not what you want in this case though.
20160224 20:46:48< celticminstrel> I'm thinking it might actually make sense for tgroup to extend tdispatcher.
20160224 20:47:08< vultraz> celticminstrel: ok, so by current design, you can either call window.add_to_group("group_id", widget) or have the widget implement a wrapper to push 'this'.
20160224 20:47:41< SigurdFD> zookeeper: huh, wonder if the behavior is related to https://gna.org/bugs/?24448 ? probably not, but who knows.
20160224 20:47:44< celticminstrel> Forget about the current design for a moment. Ideally, how would you want the group to be constructed?
20160224 20:47:56< vultraz> either of those ways
20160224 20:47:59< vultraz> whichever is simpler
20160224 20:48:11< celticminstrel> I don't understand the second way.
20160224 20:48:21< zookeeper> SigurdFD, uh... i don't see why you'd think they might be related
20160224 20:48:40< celticminstrel> Basically you're saying that constructing the group solely consists of populating it with widgets? There are no other steps to take?
20160224 20:48:48< vultraz> correct
20160224 20:48:56< vultraz> is this bad design?
20160224 20:49:13< celticminstrel> I'm not saying it is or isn't.
20160224 20:49:25< celticminstrel> So, once the group is constructed, how do you want it to be used?
20160224 20:50:06< vultraz> well, since I now have iterator pairs
20160224 20:50:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 20:50:25< celticminstrel> Forget about that for a moment. Just say how you want it to be used.
20160224 20:50:59< vultraz> ok, you should be able to loop through it
20160224 20:51:17< vultraz> and also I'd like to be able, some way, to possibly have values bound to each widget
20160224 20:51:36< vultraz> not as functionality of tgroup, of course
20160224 20:51:36< celticminstrel> If you're just looping through it, you might as well just use a vector instead.
20160224 20:51:50< vultraz> more like of tselectable_
20160224 20:52:27< SigurdFD> zookeeper: only cause they were both begining of scenario. It was silly notion.
20160224 20:52:30< celticminstrel> So, you want values bound to each widget. That tells me that you might want your add_to_group() function to take two parameters. Anything else about how it would be used?
20160224 20:53:14< vultraz> not presently, no
20160224 20:53:37< celticminstrel> Hmmm...
20160224 20:53:42< vultraz> I assume in your suggestion add_to_group is templated?
20160224 20:55:23< celticminstrel> Why?
20160224 20:55:40< celticminstrel> You might be right.
20160224 20:55:56< celticminstrel> Or, maybe not.
20160224 20:56:51< mattsc> vultraz, celticminstrel: I have moved all AI pages to be deleted here: https://wiki.wesnoth.org/Category:Candidates_for_deletion
20160224 20:57:06< vultraz> because the radio button usecase means arbitrary values 
20160224 20:57:18< mattsc> I have also removed them from the AI Category, so it does not matter how long it takes until you actually delete them.
20160224 20:57:27< celticminstrel> I'm not sure I get what you're saying.
20160224 20:57:43< celticminstrel> As an aside... as a testing ground for groups, I suggest converting a simple dialog (maybe tadvanced_graphics_options) to use them. Maybe even before starting on the actual implementation of the groups.
20160224 20:58:01< vultraz> I'm using prefs as a testing ground
20160224 20:58:06< vultraz> but yes, that's also a good idea
20160224 20:58:15< celticminstrel> Another aside: I suspect there's no need for groups to have an ID or be tracked by the twindow object.
20160224 20:58:19< vultraz> (oh, and editor edit side)
20160224 20:58:32< celticminstrel> I think a simple dialog is best for a testing ground.
20160224 20:58:44< vultraz> hm
20160224 20:58:55< vultraz> what do you mean "not tracked by the twindow object"
20160224 20:59:10< vultraz> also of course they need an id
20160224 20:59:12< celticminstrel> Thinking about it, I think add_to_group could be templated, but I think it would be easier if it wasn't.
20160224 20:59:19< vultraz> how else do you specify what group to add to?
20160224 20:59:31< celticminstrel> I mean how you had the twindow keep track of all groups in a list or whatever.
20160224 20:59:44< vultraz> hm
20160224 20:59:45< celticminstrel> As for how to specify what group to add to, if you have the group object, that's not an issue.
20160224 20:59:49< vultraz> well..
20160224 21:00:19< vultraz> but how do you get it
20160224 21:00:21< celticminstrel> If you want to add WML support for groups, then you'll need an ID and some other stuff, but if you also want the groups to be generic, I think that will be very difficult.
20160224 21:00:27< celticminstrel> How do you get the group?
20160224 21:00:29< vultraz> create a new one?
20160224 21:00:32< celticminstrel> Yeah.
20160224 21:00:46< vultraz> like, what, new foo tgroup() add_to_group(foo, this)
20160224 21:00:51< celticminstrel> No.
20160224 21:00:52< celticminstrel> Declare it as a member of the individual dialog's class.
20160224 21:01:07< celticminstrel> And use add_to_group in preshow or in the constructor.
20160224 21:01:36< celticminstrel> I don't understand why you wrote "add_to_group(foo, this)".
20160224 21:01:37< vultraz> hmmm
20160224 21:01:38< irker717> wesnoth: sigurdfdragon wesnoth:master c970efa6bf44 / data/campaigns/tutorial/ (3 files in 2 dirs): T: Make Konrad & Li'sar unnameable. https://github.com/wesnoth/wesnoth/commit/c970efa6bf44da11736430955dae6f5bf0a7cb84
20160224 21:01:40< irker717> wesnoth: sigurdfdragon wesnoth:master d99806cbbfb2 / data/campaigns/tutorial/scenarios/01_Tutorial_part_1.cfg: T: Make Delfador have same traits as HttT. https://github.com/wesnoth/wesnoth/commit/d99806cbbfb2ae86eb48f147c724a3751d134210
20160224 21:01:42< irker717> wesnoth: Charles Dang wesnoth:master 9358af7ad12b / data/campaigns/tutorial/ (3 files in 2 dirs): Merge pull request #609 from sigurdfdragon/Tutorial_Fixes_2016_02_24 https://github.com/wesnoth/wesnoth/commit/9358af7ad12b0a251792fdf4b8d48677e899dde6
20160224 21:01:51< vultraz> celticminstrel: pseudocode
20160224 21:02:02< celticminstrel> Okay fine whatever.
20160224 21:02:06< celticminstrel> Very pseudo.
20160224 21:02:09< vultraz> that actually makes sense
20160224 21:02:20< vultraz> keep groups on a per-dialog basis
20160224 21:02:39< celticminstrel> When I said that perhaps add_to_group should not be templated, the alternative would be moving that template argument to the class.
20160224 21:02:50< vultraz> this would greatly simplify things 
20160224 21:02:59< vultraz> since then I don't need a map of groups in the window
20160224 21:03:14< vultraz> which is tbh, only used internally because of group id
20160224 21:03:46< vultraz> (basically the map is just used in order to check if a group with that id exists)
20160224 21:03:59< vultraz> but if we keep tgroup objects in the dialog classes
20160224 21:04:22< vultraz> we remove the twindow interaction layer
20160224 21:05:07< vultraz> so all we'd have is tgroup, then dialog classes with tgroup
20160224 21:05:12< vultraz> and tgroup already has add_member
20160224 21:05:42< vultraz> so we'd just do like, group.add_member(find_widget<blablabla>())
20160224 21:05:55< vultraz> celticminstrel: how's that sound 
20160224 21:05:57< celticminstrel> Something along those lines, yeah.
20160224 21:06:09< celticminstrel> I mentioned maybe adding a second parameter there.
20160224 21:06:15< vultraz> right
20160224 21:06:37< celticminstrel> Alternatively, you could have it call a function to determine the value associated with the widget.
20160224 21:06:51< vultraz> I can change the member vector to a vector of pairs 
20160224 21:06:55< celticminstrel> I think that might be harder, though.
20160224 21:07:03< vultraz> except...
20160224 21:07:23< celticminstrel> What I said might be harder, I mean. Not the vector of pairs.
20160224 21:07:24< vultraz> can you even have a vector of pairs where the second value is arbitrary?
20160224 21:07:34< celticminstrel> Define "arbitrary".
20160224 21:07:48< vultraz> like, tgroup would have no idea what the values' types would be
20160224 21:07:59< celticminstrel> If that's really what you want, then use boost::any.
20160224 21:08:05< celticminstrel> Is it really what you want, though?
20160224 21:08:07< vultraz> so you;d have std::vector<std::pair<twidget*, ?? > >
20160224 21:08:19< celticminstrel> In practice I suspect that all the values would be of the same type.
20160224 21:08:25< vultraz> yes
20160224 21:08:29< vultraz> but what type would that be
20160224 21:08:40< celticminstrel> You don't know.
20160224 21:08:43< celticminstrel> So template it.
20160224 21:09:09< celticminstrel> (And if there was some outlying case where the values really could be of any type, you could just assign boost::any as that type.)
20160224 21:09:20< vultraz> class template it?
20160224 21:09:38< celticminstrel> If you function template it, you're not solving that problem - they can still be of any type.
20160224 21:09:40< celticminstrel> So yeah.
20160224 21:09:52< vultraz> ok so
20160224 21:09:55< vultraz> then I would uh
20160224 21:10:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 21:11:08< vultraz> ok so I'd have template<class T> class tgroup { private : std::vector<std::pair<twidget*, T> > members_ }, then in the dialog classes, tgroup<int> or something, which would make the vector a vector of widget/int pairs?
20160224 21:11:45< celticminstrel> Hmm, I suppose you could do that.
20160224 21:11:59< celticminstrel> That would work as you say.
20160224 21:12:24< vultraz> good, good
20160224 21:12:34< celticminstrel> This is a heterogenous container, of course.
20160224 21:12:42< celticminstrel> Rather than a homogenous one.
20160224 21:12:42< vultraz> one thing - what if I don't want a value associated with the widget
20160224 21:13:09< celticminstrel> I guess you could specialize it for void.
20160224 21:13:12< vultraz> can I set the second pair value to NULL? Maybe it should be T*
20160224 21:13:21< celticminstrel> We can get to that if you actually find you need it.
20160224 21:13:28< celticminstrel> Don't go making it T*.
20160224 21:13:55< celticminstrel> It would be possible to set it up so that you can do tgroup<void> if you want no value.
20160224 21:14:06< vultraz> hm
20160224 21:14:28< vultraz> ok
20160224 21:14:36< vultraz> well let me go implement this 
20160224 21:14:56< celticminstrel> Just one question. You're back to using twidget there. Any particular reason?
20160224 21:15:14< vultraz> because...er...
20160224 21:15:23< vultraz> no reason, really...
20160224 21:15:46< vultraz> specializing tgroup to selectable widgets means I could also have it implement a get_active_widget member
20160224 21:15:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 276 seconds]
20160224 21:16:27< vultraz> as opposed to making that virtual and 
20160224 21:16:29< vultraz> er
20160224 21:16:31< vultraz> well
20160224 21:16:42< vultraz> I suppose I could keep it twidget and declare group a friend of tselectable
20160224 21:17:06< vultraz> what do you say
20160224 21:17:35< celticminstrel> Well, the only use-case I know of for groups is homogenous - a group of toggle buttons.
20160224 21:17:49< vultraz> yes, true
20160224 21:17:54< vultraz> I guess I'll keep it simple for now
20160224 21:17:54< celticminstrel> Actually, I don't know of any use-case that's not toggle-buttons.
20160224 21:18:13< vultraz> let me get on this
20160224 21:20:07< vultraz> ok, so am I allowed to pass value by reference to add_member?
20160224 21:20:13< vultraz> ie, void add_member(tselectable_* widget, T& value);
20160224 21:20:52< celticminstrel> I don't see why not. You could even go const reference.
20160224 21:21:03< celticminstrel> You're probably going to copy it to the vector, anyway, right?
20160224 21:21:18< vultraz> yes
20160224 21:21:33< celticminstrel> BTW, there exists std::make_pair(a, b). Use it.
20160224 21:21:46< vultraz> I am aware
20160224 21:21:52< celticminstrel> Ah, okay.
20160224 21:22:12< vultraz> ok, so since we're dropping group ids, that means I don't need a constructor anymore
20160224 21:22:55< celticminstrel> Not as far as I know. It can always be added back later if a need comes up.
20160224 21:23:15< vultraz> this is getting to the point where it might as well be a header-only class
20160224 21:23:26< celticminstrel> It should be a header-only class no matter what.
20160224 21:23:33< celticminstrel> Because it's templated.
20160224 21:24:25< celticminstrel> You can define the functions inline within the class definition or below the main class definition, but they need to be in the header.
20160224 21:24:50< celticminstrel> (When templates are split into two files, usually the "source" file is included by the header.)
20160224 21:25:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 21:26:43< vultraz> celticminstrel: ok, how's this look http://pastebin.com/nw63HQJh
20160224 21:28:02< celticminstrel> Looks okay for now (though I expect it won't compile).
20160224 21:28:53< mattsc> Ugh — and now all the links in the sidebars are updated too  :P
20160224 21:29:02< vultraz> out of curiosity
20160224 21:29:03< vultraz> could 
20160224 21:29:05< vultraz> members_.erase(std::find(members_.begin(), members_.end(), widget));
20160224 21:29:07< celticminstrel> ma: Yay!
20160224 21:29:07< vultraz> be changed to
20160224 21:29:19< vultraz> members_.erase(std::find(group_iterator, widget));
20160224 21:29:55< celticminstrel> vultraz: No, I don't believe the standard library functions support ranges.
20160224 21:30:14 * celticminstrel assumes you mean widgets() in place of group_iterator, which is a type.
20160224 21:30:40< celticminstrel> Also, find() won't work there at all. You need either find_if or manual iteration.
20160224 21:30:49< vultraz> eh?
20160224 21:31:04< mattsc> celticminstrel: with that I have two major sets of modifications left: Creating_Custom_AIs and Wesnoth_AI; and then, of course, I need to go over everything with a fine comb and clean it up and cross-link correctly.
20160224 21:31:24< mattsc> But that’s much better than it used to be!
20160224 21:31:26< celticminstrel> The type of a value in the list is pair<tselectable_*, T>. Therefore, searching for a value of type tselectable_* in the list won't work.
20160224 21:31:36< vultraz> dammit
20160224 21:31:40< vultraz> what is this find_if?
20160224 21:31:47< celticminstrel> Google it.
20160224 21:31:51< celticminstrel> "c++ find_if"
20160224 21:32:06< celticminstrel> The first result or two should be documentation.
20160224 21:32:37< celticminstrel> But since we're not using C++11, you would need to define the predicate function separately if you used find_if.
20160224 21:32:48< celticminstrel> In C++11 you'd use a lambda expression.
20160224 21:33:05< vultraz> :|
20160224 21:33:12 * vultraz curses
20160224 21:33:45< celticminstrel> It's not hard to do.
20160224 21:33:55< celticminstrel> It's also not hard to iterate manually.
20160224 21:34:05< celticminstrel> Your choice.
20160224 21:34:15< celticminstrel> I doubt get_widgets() is required.
20160224 21:34:36< vultraz> probably not
20160224 21:35:32< vultraz> hm
20160224 21:35:46< vultraz> do you think we need to keep the 'widget keeps pointer to its group' thing?
20160224 21:36:18< celticminstrel> Well, that all depends on how you intend to set up the group action.
20160224 21:36:44< celticminstrel> It can probably be omitted.
20160224 21:36:58< vultraz> right...group action
20160224 21:37:20< vultraz> right, I'm specializing this for tselectable 
20160224 21:38:21< vultraz> somehow I still feel... maybe the group actions and get_active_member might be better in tselectable
20160224 21:38:52< vultraz> the actions, at least
20160224 21:38:59< vultraz> but the event needs to be actually fired somewhere..
20160224 21:39:06< vultraz> do I need to hook into tdispatcher? :|
20160224 21:40:10< celticminstrel> Possibly?
20160224 21:40:18< vultraz> currently, for toggle buttons at least, the actions are dealt with in the left click event
20160224 21:41:09< vultraz> I think we *might* need to keep that group pointer
20160224 21:41:11-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20160224 21:41:22< celticminstrel> I think it would be great if you could do something like connect_signal_notify_modified(my_group, ...)
20160224 21:41:42-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160224 21:41:45< celticminstrel> Which would mean that the individual widgets in the group would need to trigger the group to do something.
20160224 21:41:55< celticminstrel> Which means they need a pointer back to the group.
20160224 21:42:03< celticminstrel> However, that pointer could be stored as part of a bound function.
20160224 21:42:05< celticminstrel> Probably.
20160224 21:43:18< vultraz> on a per-dialog basis, or in tselectable 
20160224 21:44:42< vultraz> hm
20160224 21:44:55< vultraz> keeping the group pointer in tselectable means providing a template argument
20160224 21:45:28 * vultraz comments out for now
20160224 21:50:22< vultraz> celticminstrel: the simplest solution for the event thing is to keep the group pointer, then in the left-click event for a specific widget, check if it's in the group, then if so, fire some function that gets the iterator from the group and loops over, executing a specific action
20160224 21:50:31< vultraz> but I'm not sure that's the best solution
20160224 21:51:00< celticminstrel> Okay, let me see if I understood what you just said.
20160224 21:51:20< celticminstrel> Each widget in the group has a left-click handler.
20160224 21:51:38< vultraz> yes
20160224 21:51:45< celticminstrel> That left-click handler loops through the group and unsets all other widgets in the group, then makes sure the triggered widget is set.
20160224 21:52:06< vultraz> basically, that, yes
20160224 21:52:18< celticminstrel> Personally, I think that that left-click handler is best implemented as a member function in ttoggle_button, which can then be bound by the tgroup.
20160224 21:52:34< vultraz> but what if the dialog wanted to set a custom callback function...
20160224 21:52:40< celticminstrel> The tgroup would be a paremeter to the function.
20160224 21:52:54< celticminstrel> If the dialog wanted a custom callback function, I dunno.
20160224 21:53:10< vultraz> you'd need the dispatcher
20160224 21:53:22< vultraz> anyway...er
20160224 21:53:24< celticminstrel> If all they want is to know when the selection changes, mind you, it would make more sense to bind an event to the group itself.
20160224 21:53:42< vultraz> in tgroup?
20160224 21:53:46< vultraz> but *how*
20160224 21:53:52< vultraz> I don't understand the gui2 event chain at all
20160224 21:53:55-!- louis94 [~~louis94@91.178.241.231] has quit [Ping timeout: 252 seconds]
20160224 21:54:21< celticminstrel> I don't know exactly, but I'm pretty sure inheriting from tdispatcher is required, at least.
20160224 21:54:33< celticminstrel> If you're lucky, maybe that's all that's required.
20160224 21:55:28< vultraz> well
20160224 21:55:41< vultraz> hm
20160224 21:56:03< vultraz> I *could* have a connect_signal_notify_modified event for every widget in the group
20160224 21:56:40< vultraz> but does that work with toggle-type widgets?
20160224 21:56:45< vultraz> I don't think so, actually..
20160224 21:56:53< vultraz> errrr...
20160224 21:56:55< vultraz> hmmm
20160224 21:57:44< celticminstrel> I suggest you first get it working without a possibility for custom callbacks or notify-modified callbacks.
20160224 21:58:29< vultraz> well in some way I need to have that event fired
20160224 21:58:34< vultraz> or else it dosn't work at all
20160224 21:59:57< celticminstrel> You only need notify-modified if a dialog wants to react to a change in the selected value.
20160224 22:00:17< celticminstrel> If it's just a simple selection, then all you need is the basic left-click behaviour.
20160224 22:00:20< celticminstrel> So get that working first.
20160224 22:02:57< vultraz> true
20160224 22:03:03< vultraz> i guess my suggestion above will do
20160224 22:03:06< vultraz> but
20160224 22:03:27< vultraz> how do I keep a tgroup pointer if i need to specify a template value
20160224 22:03:36< vultraz> if the template value is determined in the dialogs?
20160224 22:03:49< celticminstrel> Don't keep a pointer. Bind it into the callback function.
20160224 22:04:15< celticminstrel> Come to think of it, the callback function would probably need to be templated too in order for that to work.
20160224 22:04:27< vultraz> grrrrrr
20160224 22:04:35< vultraz> ttoggle_button already has a signal_handler_left_button_click function
20160224 22:04:48< celticminstrel> Assuming of course that it's a generic callback function.
20160224 22:05:10< celticminstrel> What is signal_handler_left_button_click?
20160224 22:05:11< vultraz> which is private
20160224 22:05:20< vultraz> it's bound to
20160224 22:05:28< vultraz> event::LEFT_BUTTON_CLICK
20160224 22:05:45< celticminstrel> It's something bound using connect_signal_?
20160224 22:05:50< vultraz> yes
20160224 22:05:54< celticminstrel> What does it do?
20160224 22:06:10< vultraz> it handled clicking on a toggle button
20160224 22:06:12< vultraz> but
20160224 22:06:14< vultraz> er
20160224 22:06:15< vultraz> hm
20160224 22:06:20< celticminstrel> So basically, it toggles its state?
20160224 22:06:34< celticminstrel> I wonder what happens when you bind two callbacks to the same signal on the same widget...
20160224 22:06:40< vultraz> yes, and fires callbacks
20160224 22:06:50< vultraz> which I just remembered, dialogs can do
20160224 22:06:52< celticminstrel> EH?h
20160224 22:06:56< celticminstrel> ^Eh
20160224 22:06:58< vultraz> connect_signal_mouse_left_click
20160224 22:07:07< celticminstrel> I'm confused now.
20160224 22:07:11< vultraz> yes
20160224 22:07:21< vultraz> this is gui2 for you
20160224 22:07:26< vultraz> i don't envy Aginor  :P
20160224 22:07:42< celticminstrel> Or maybe it's just your explanations which are bad.
20160224 22:07:58< celticminstrel> Okay, so there's a "state change" callback.
20160224 22:08:03< vultraz> basically, connect_signal_mouse_left_click is a function used in dialogs to bind functions to clicks on widgets
20160224 22:08:07-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20160224 22:08:27< celticminstrel> When I look at the source myself it seems pretty straightforward.
20160224 22:08:28< vultraz> and this is *different* from the internal ttoggle_button::signal_handler_left_button_click function
20160224 22:08:45< vultraz> which is different from the virtual tselectable_::set_callback_state_change function
20160224 22:08:54< celticminstrel> I dunno about that.
20160224 22:08:58< vultraz> which is unimplemented in ttoggle_button
20160224 22:09:17< celticminstrel> It looks like signal_handler_left_button_click is bound using connect_signal_mouse_left_click.
20160224 22:09:42< vultraz> er...no..
20160224 22:09:44< vultraz> ?
20160224 22:09:55< vultraz> or
20160224 22:09:57< vultraz> hm
20160224 22:10:01< vultraz> actually it kinda is
20160224 22:10:01< celticminstrel> Well, it called connect_signal directly, but same effect.
20160224 22:10:11< vultraz> yes
20160224 22:10:13< vultraz> ok
20160224 22:10:16< celticminstrel> (connect_signal_left_button_click simply forwards to connect_signal.)
20160224 22:11:32< celticminstrel> It also looks like you could connect a signal so as to ensure that it occurs before signal_handler_left_button_click,
20160224 22:11:32< vultraz> so somewhere I need to hook into the dispatcher with a function that takes a group as an argument
20160224 22:12:06< celticminstrel> You'd do togglebutton.connect_signal<event::LEFT_BUTTON_CLICK>(callback_function, front_child)
20160224 22:12:37< vultraz> I know about the child callback order, yes
20160224 22:12:44< celticminstrel> For now, the callback function could be something like &ttoggle_button::exclude_group_members
20160224 22:13:11< celticminstrel> Which would be defined to take the tgroup as an argument and would deselect all members of the group.
20160224 22:13:25< celticminstrel> (You'd pass the group in via boost::bind)
20160224 22:13:52< vultraz> shoudn't that be tselectable behavior, though?
20160224 22:13:55< celticminstrel> And if set up right, exclude_group_members would be called first, followed by signal_handler_left_button_click, and everything would work correctly.
20160224 22:14:00< vultraz> maybe I can implement it there 
20160224 22:14:15< vultraz> then bind it to the signal handler in ttoggle_button
20160224 22:14:17< celticminstrel> It can't be tselectable_ behaviour in my opinion, because it assumes that each widget can only have one of two values.
20160224 22:14:39< celticminstrel> Though, you could make it set to 0 or something, I suppose.
20160224 22:14:51< celticminstrel> Then I guess it could be placed in tselectable_, if you really want.
20160224 22:15:12< celticminstrel> I was assuming tgroup would handle the signal binding.
20160224 22:16:43< vultraz> ok so, in tgroup::add_member
20160224 22:16:45< vultraz> something like
20160224 22:17:48< vultraz> connect_signal<event::LEFT_BUTTON_CLICK>(boost::bind(&tselectable_::group_actions, this));
20160224 22:17:49< vultraz> ?
20160224 22:18:06< vultraz> front_child, of course
20160224 22:18:12< celticminstrel> Well, widget.connect_signal
20160224 22:18:24< celticminstrel> That basically looks fine.
20160224 22:18:27< vultraz> ok
20160224 22:18:31< vultraz> I think that will work
20160224 22:18:33< vultraz> let me try
20160224 22:19:40-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20160224 22:23:35< vultraz> celticminstrel: don't we want 'this, _2'?
20160224 22:24:30< celticminstrel> Hmm.
20160224 22:25:00< celticminstrel> It's a member function of tselectable_, so the first argument needs to be a pointer to that.
20160224 22:25:21< celticminstrel> So "this" only works if you're calling it from a tselectable_, which I believe you're not.
20160224 22:25:46< vultraz> I am not
20160224 22:26:21< vultraz> or
20160224 22:26:30< vultraz> hm
20160224 22:26:48< vultraz> I was thinking an argument could be passed somewhere to tgroup with a function to bind to its members
20160224 22:26:54< vultraz> but not sure where
20160224 22:27:21-!- gfgtdf [~chatzilla@f054061068.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 44.0.2/20160210153822]]
20160224 22:27:43< celticminstrel> I thought of that too. That wouldn't be very difficult to add once you have the basic setup working.
20160224 22:28:32< vultraz> celticminstrel: a private boost::function<void(tgroup*)> pointer?
20160224 22:28:45< vultraz> that would make a lot of sense, actually
20160224 22:28:58< celticminstrel> I suggest you get it working first.
20160224 22:29:11< celticminstrel> But that's more or less on the right track. It's not right, but it's the right idea.
20160224 22:29:17< vultraz> again, can't get it working without this :P
20160224 22:29:34< celticminstrel> Yes you can, you can hardcode a specific callback function.
20160224 22:29:51< vultraz> I thought you just said I can't
20160224 22:29:54< vultraz> [09:25:19]	celticminstrel	So "this" only works if you're calling it from a tselectable_, which I believe you're not.
20160224 22:30:02< celticminstrel> ...
20160224 22:30:04< vultraz> the 'this' was for the tgroup object as an argument 
20160224 22:30:13< vultraz> which is why I asked if we needed _2
20160224 22:30:24< celticminstrel> I'm not sure what _2 is.
20160224 22:30:30< celticminstrel> Or _1 for that matter.
20160224 22:30:38< vultraz> unless you're saying i should do '&widget, this'?
20160224 22:31:15< celticminstrel> Okay, so in this specific case (binding a signal) _1 is the widget, _2 is the event, _3 and _4 are booleans which we almost certainly don't care about.
20160224 22:31:35< celticminstrel> However, _1 is a reference to the widget, so yeah, "&widget, this" will have to do.
20160224 22:32:05< celticminstrel> I don't think you can put a reference instead of a pointer in as the first argument to a bound member function.
20160224 22:32:15< vultraz> so:
20160224 22:32:18< vultraz> 		widget.connect_signal<event::LEFT_BUTTON_CLICK>(
20160224 22:32:20< vultraz> 			boost::bind(&tselectable_::group_operator, &widget, this), front_child);
20160224 22:32:27< celticminstrel> Looks good.
20160224 22:34:15< celticminstrel> For the record, tselectable_ is literally either a toggle button or a toggle panel. Or, for some reason, a combobox.
20160224 22:34:27-!- louis94 [~~louis94@91.178.241.231] has joined #wesnoth-dev
20160224 22:37:36< vultraz> hm.. I don't think widget->connect_signal is right
20160224 22:37:37-!- Appleman1234 [~Appleman1@KD119104016245.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20160224 22:37:43< vultraz> just 'connect_signal', I think
20160224 22:38:50< celticminstrel> No, that would connect the signal to the tgroup.
20160224 22:39:10< celticminstrel> (If tgroup even inherited from tdispatcher in the first place.)
20160224 22:39:23< celticminstrel> connect_signal is a member function defined in tdispatcher.
20160224 22:39:31< vultraz> but it says tselectable doesn't have a connect_signal member :|
20160224 22:39:48< celticminstrel> Ah, that's because tselectable_ doesn't inherit from tdispatcher.
20160224 22:39:53-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has quit [Read error: No route to host]
20160224 22:40:01< celticminstrel> A dynamic_cast to tdispatcher should work.
20160224 22:40:39< celticminstrel> Or, I guess, you could use twidget instead of tselectable_.
20160224 22:41:03< celticminstrel> Or, option three would be for the widget class to be a template argument to the tgroup.
20160224 22:44:22-!- Kwandulin [~Miranda@p200300760F0BC5832CCC877D7A01ED7D.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160224 22:44:32< vultraz> ok why the hell do I keep getting errors about stuff not being declared even though the headers are being included
20160224 22:45:14< celticminstrel> I have no idea!
20160224 22:45:16< vultraz> maybe it's because both tselectable_.hpp and group.hpp include each other?
20160224 22:45:35< celticminstrel> I doubt that.
20160224 22:45:52< celticminstrel> They shouldn't include each other, of course.
20160224 22:46:17< vultraz> why?
20160224 22:46:57< celticminstrel> It should be obvious why.
20160224 22:47:27< vultraz> but if I don't I get errors about incomplete types in both :|
20160224 22:47:33< celticminstrel> What detail requires the inclusion of selectable.hpp?
20160224 22:47:37< celticminstrel> The ::?
20160224 22:47:48< vultraz> the &tselectable_::group_operator bit
20160224 22:47:59< celticminstrel> Figures.
20160224 22:49:03< celticminstrel> And what detail requires the inclusion of group.hpp in selectable.hpp?
20160224 22:49:28< vultraz> 	void group_operator(tgroup* group)
20160224 22:49:29< vultraz> 	{
20160224 22:49:31< vultraz> 		FOREACH(AUTO& widget, group->widgets())
20160224 22:49:46< celticminstrel> Is there a selectable.cpp?
20160224 22:50:01< vultraz> no
20160224 22:50:40< celticminstrel> Figures. tselectable_ was supposed to be an interface, I think.
20160224 22:50:47< celticminstrel> So it wouldn't need a source file normally.
20160224 22:51:37< celticminstrel> As a temporary measure, you could declare it in selectable.hpp but define it in some other file, like toggle_button.cpp.
20160224 22:52:02< celticminstrel> Without the definition of that function, a forward definition of tgroup would be sufficient.
20160224 22:52:22< celticminstrel> (When forward-declaring a template, you also need to include the "template" part of the declaration.)
20160224 22:52:31< celticminstrel> (Just to be clear.)
20160224 22:53:29< vultraz> wouldn't to have to be virtual, then?
20160224 22:55:31< celticminstrel> No.
20160224 22:56:28< celticminstrel> The reason I'm suggesting such a temporary method is because I'm thinking of a way to avoid referencing tselectable_ from group.hpp
20160224 22:56:42< celticminstrel> However, I have to go cook now, so back in awhile.
20160224 22:57:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160224 22:58:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160224 23:02:49-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160224 23:04:11-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-176-151.zoominternet.net] has joined #wesnoth-dev
20160224 23:07:49< SigurdFD> vultraz: Do any of my pr's need changelog entries? I'm thinking #601 commit 1 & #592, maybe.
20160224 23:08:40-!- horrowind [~Icedove@2a02:810a:8380:834:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160224 23:35:03-!- Appleman1234 [~Appleman1@KD119104005043.au-net.ne.jp] has joined #wesnoth-dev
20160224 23:43:15-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 250 seconds]
20160224 23:55:30-!- louis94 [~~louis94@91.178.241.231] has quit [Quit: Konversation terminated!]
20160224 23:58:42-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit []
20160224 23:59:24-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
--- Log closed Thu Feb 25 00:00:36 2016