--- Log opened Mon Aug 15 00:00:35 2016
20160815 00:02:22< vultraz> gaaaahHHH WNARROWING
20160815 00:02:47< shadowm> Any news on the transparency situation?
20160815 00:03:04< vultraz> I haven't had time to pursue it
20160815 00:03:08< vultraz> Yet
20160815 00:03:20< shadowm> Is this going to become the next GUI2 then?
20160815 00:03:35< vultraz> :|
20160815 00:03:41< shadowm> "I don't have time to attend to something that is obviously urgent and it's not anyone else's business so leave it alone and wait for me."
20160815 00:06:25 * vultraz curses wnarrowing
20160815 00:06:54< vultraz> wonder if I should just make the constructor unsigned
20160815 00:07:26< celmin> Randomly switching between signed and unsigned is not a good idea.
20160815 00:08:09< vultraz> guess i have to use tpoint(a,b) explicitely
20160815 00:10:10< vultraz> seems it doesn't like the {} syntax with this
20160815 00:10:12< vultraz> dunno why
20160815 00:10:26< celmin> It's complicated.
20160815 00:15:35-!- irker658 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160815 00:16:17< gfgtdf> vultraz: you coudl make add a second thor construicto of tpaoiint that takes unsigned ints, not sure if that would work.
20160815 00:16:22-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160815 00:33:28< celmin> Why'd you feel the need to emphasize the lack of an official Facebook page?
20160815 00:33:35< celmin> (Just curious, not complaining.)
20160815 00:34:56-!- irker310 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160815 00:34:56< irker310> wesnoth: Charles Dang wesnoth:master ddc8d84c061b / src/gui/ (21 files in 4 dirs): Made tpoint default-initializeable and cleaned up its uses https://github.com/wesnoth/wesnoth/commit/ddc8d84c061b225d3a4b20082ef1a3c24693f3d4
20160815 00:40:23-!- gfgtdf [~chatzilla@x4e36ad0f.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20160815 00:50:04-!- Bonobo [~Bonobo@2001:44b8:254:3200:14d2:cac6:2190:470c] has quit [Ping timeout: 260 seconds]
20160815 00:50:59-!- Bonobo [~Bonobo@2001:44b8:254:3200:407:dc4d:a335:7099] has joined #wesnoth-dev
20160815 00:53:32< vultraz> at some point I'd like to do something about the fact that you keep having to pass references to twindow around when you do dialogs
20160815 00:53:51< vultraz> there's only one damn window
20160815 00:54:17< celmin> Not really?
20160815 00:54:28< celmin> Anyway, the window is the toplevel widget that holds all the others.
20160815 00:56:46< vultraz> not right now, though.
20160815 00:57:00< vultraz> there're better things to clean up
20160815 00:58:25< vultraz> need to properly set up settings
20160815 00:58:31< vultraz> for mp create..
20160815 00:58:39< vultraz> then figure out how to actually start a game
20160815 00:58:47< Aginor> shadowm: presumably
20160815 00:59:38< Aginor> there should probably be a discussion about the things we are seeking clarification about and the chosen delivery method
20160815 00:59:43< Aginor> and assign actions to people
20160815 00:59:46< Aginor> and follow up
20160815 00:59:55< Aginor> like any other meeting
20160815 00:59:58< Aginor> etc
20160815 01:00:06< celmin> vultraz: Where do you need to pass references to twindow around?
20160815 01:00:12< vultraz> I don't envy myself the task of getting dave to sit down and talk
20160815 01:00:20< vultraz> celmin: every time I have a function that uses find_widget
20160815 01:00:35< celmin> Well, you certainly need a twindow there.
20160815 01:00:43< Aginor> vultraz: that's why shadowm's suggestion of a letter from all of us makes sense
20160815 01:00:50< celmin> So what do you propose to use instead?
20160815 01:01:08< vultraz> I dunno
20160815 01:01:13< vultraz> maybe some get_window() methid
20160815 01:01:13< celmin> ^do
20160815 01:01:14< vultraz> method
20160815 01:01:27< celmin> But a dialog does not logically have a window.
20160815 01:01:35< celmin> It only has a window while executing show().
20160815 01:01:35< vultraz> this is true
20160815 01:01:45< celmin> I suppose if the method were protected it might be okay, but I dunno...
20160815 01:02:02< celmin> It's still true that it doesn't have one at all times.
20160815 01:02:33< celmin> I don't really see the problem with passing around a twindow, anyway.
20160815 01:04:59< vultraz> another thing we should do is allow more use of lambdas
20160815 01:05:17-!- enchi [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 258 seconds]
20160815 01:07:57< vultraz> purge stuff like dialog_callback
20160815 01:13:56< vultraz> (this https://github.com/wesnoth/wesnoth/blob/master/src/gui/dialogs/helper.hpp )
20160815 01:14:16-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160815 01:28:51-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 01:28:52< travis-ci> wesnoth/wesnoth#10403 (master - ddc8d84 : Charles Dang): The build has errored.
20160815 01:28:52< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152286261
20160815 01:28:52-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 01:45:22-!- oldlaptop [~quassel@50-37-63-133.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20160815 01:57:05-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160815 01:58:36< celmin> So vultraz's latest commit adds another instance of the erroring line...
20160815 01:58:50< vultraz> what line?
20160815 01:58:58< celmin> Maybe I should have asked if ancestral can build master. It could be something not worth fixing.
20160815 01:59:40< celmin> The add_node("whatever", {}) lines
20160815 02:16:02-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20160815 02:21:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20160815 02:24:28-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160815 02:33:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160815 02:42:36< aeth> Apparently there's a library called SDL-gpu. https://news.ycombinator.com/item?id=12287389
20160815 02:42:48< aeth> This might be better than porting entirely over to OpenGL, or at least would be a good transition
20160815 02:43:48< Aginor> aeth: we know
20160815 02:44:07< Aginor> but thanks :)
20160815 02:45:07< aeth> I would be willing to try to write a water/lava shader, and lighting shaders.
20160815 02:47:26-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160815 02:54:27< celmin> I didn't realize that SDL_GPU was actually an external library, since Wesnoth includes it within the repo.
20160815 02:55:00-!- oldlaptop [~quassel@50-37-63-133.mskg.mi.frontiernet.net] has quit [Ping timeout: 276 seconds]
20160815 02:55:03< aeth> I didn't realize that Wesnoth was using SDL_GPU because Wesnoth includes it within the repo itself
20160815 02:55:06< aeth> rather than as a dependency
20160815 02:55:11< aeth> I know Wesnoth's dependencies
20160815 02:55:12< celmin> It's not actually using it.
20160815 02:55:15< aeth> ah
20160815 02:55:22-!- enchi [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160815 02:55:28< celmin> At least, not on any default builds.
20160815 02:55:43< vultraz> the code hasn't been maintained
20160815 02:55:46< celmin> There're defines that enable it, though I suppose they probably don't compile anymore.
20160815 02:55:48< vultraz> and Aginor said he'd remove it
20160815 02:56:58< celmin> I wonder if this would be worth using for my own projects. I like SFML better, but its keyboard handling is pretty bad.
20160815 03:01:12< tad_> The unit create dialog, recruit dialog, and recall dialogs all have huge, ugly pixelated images of the units on them.
20160815 03:01:23< vultraz> deliberate
20160815 03:01:44-!- oldlaptop [~quassel@50-37-63-133.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20160815 03:01:46< celmin> I agree that it's ugly.
20160815 03:01:50< tad_> But not permanent?
20160815 03:01:59< celmin> I'm pretty sure I disagreed when you first suggested it, too.
20160815 03:02:12< celmin> I hope it's not permanent.
20160815 03:02:19< vultraz> uh...
20160815 03:02:20< vultraz> what...
20160815 03:02:26< vultraz> oh, it was zookeeper who OKed it
20160815 03:02:27< vultraz> well
20160815 03:02:42< vultraz> give me a bloody solution to the large sprite problem :|
20160815 03:02:46< celmin> Oh, did he?
20160815 03:02:55< vultraz> yes
20160815 03:03:06< celmin> I'm not sure I really get the large sprite problem.
20160815 03:04:00< tad_> is there a problem with using the size it would appear on the map?
20160815 03:04:33< celmin> I don't remember if he explained the problem clearly.
20160815 03:06:10< vultraz> the fire dragon is too small scaled to 72x72
20160815 03:07:07< tad_> Ah. So the issue is some sprites are 72x72 and some are larger?
20160815 03:07:18< vultraz> yes
20160815 03:07:35< vultraz> scaling everything to 144x144 is the compromise i came up with
20160815 03:07:57< celmin> You can't scale only if it's too large?
20160815 03:07:58< tad_> Is the max known? If so, reserve the max size and center, unscaled, would be how I'd approach it.
20160815 03:08:15< celmin> No, that'd be too big, I think.
20160815 03:08:20< vultraz> you make no sense, celmin 
20160815 03:08:23< celmin> Besides, I don't think there's a logical max.
20160815 03:08:31< celmin> vultraz:  I mean, scale down but never up.
20160815 03:08:44< vultraz> i suppose one could
20160815 03:08:56< tad_> So pick one, downsample if it's too large, and center without rescaling if it's smaller.
20160815 03:09:07< vultraz> i still need to reserve 144x144 of space
20160815 03:09:11< celmin> Why?
20160815 03:09:26-!- oldlaptop [~quassel@50-37-63-133.mskg.mi.frontiernet.net] has quit [Ping timeout: 250 seconds]
20160815 03:09:33< vultraz> because gu2
20160815 03:09:38< vultraz> gui2
20160815 03:09:44< celmin> Why?
20160815 03:09:46< tad_> Reserving a constant space is good. Keeps the stuff from jumping around as you scroll through the units, I'd guess.
20160815 03:09:49< vultraz> er, wait
20160815 03:09:51< vultraz> hm
20160815 03:09:58< vultraz> there appears to be an issue with them on master
20160815 03:10:15< vultraz> tad_: do you see sprites vertically distorted?
20160815 03:10:59< vultraz> it should look like this https://drive.google.com/file/d/0B-mR9s8FduLLSGhQbDRWMGpPTjQ/view?usp=sharing
20160815 03:11:03< vultraz> if it does not, it's a bug
20160815 03:11:26< tad_> Comparing the larger (yucky) image to the smaller one just next to it, I'd say there's some more scaling in the horizontal than vertical.
20160815 03:11:47< vultraz> right, i meant vertically squashed
20160815 03:12:40< tad_> yes, looks like it.
20160815 03:12:50< tad_> Sorry./
20160815 03:12:54< vultraz> and it does NOT look like the screenshot I posted?
20160815 03:12:59< tad_> YEs, looks liek it's vertically smashed
20160815 03:13:05< vultraz> ok
20160815 03:13:07< vultraz> that's a bug
20160815 03:13:25< vultraz> likely introduced here  https://github.com/wesnoth/wesnoth/commit/05340bb81f661806fb6cc853be292738fabda1fe
20160815 03:13:26< tad_> And I got spam on my console. Want a sample? I'ts a mess
20160815 03:13:50< celmin> The wose actually looks pretty good in that screenshot… o.o
20160815 03:13:56< celmin> Maybe cause it's non-humanoid?
20160815 03:14:18< vultraz> the wose looks especially good, yes
20160815 03:14:30< tad_> 20160814 21:21:08 error gui/layout: tgrid [] place: Failed to place a grid, we have 131,16 space but we need 139,16 space. This happened at a grid with the id '' in a 'N4gui215ttree_view_nodeE' with the id '' in a 'N4gui210ttree_viewE' with the id 'tree_details' in a 'N4gui25tgridE' with the id '' in a 'N4gui210tgeneratorINS_6policy17minimum_selection5tnoneENS1_17maximum_selection9tinfiniteENS1_9placement12tindependentENS1_13selec
20160815 03:14:54 * vultraz curses
20160815 03:14:59< celmin> I think that often corresponds with mysteriously-missing elements.
20160815 03:15:10< tad_> Not sure I can get a screen shot. May not have anything to do it installed.
20160815 03:15:52< celmin> Wesnoth has a built-in screenshot key, I think, though… not sure if it works with a dialog onscreen.
20160815 03:16:25< celmin> Actually, there are two - one renders the map, and one the window.
20160815 03:16:37< vultraz> ok, I think I know why this is happenig
20160815 03:16:42< vultraz> but how to fix it..
20160815 03:17:57< vultraz> ok
20160815 03:18:15< vultraz> because i set the w/h keys in the image widget definition
20160815 03:18:29< vultraz> it sets the w/h to the width and height of the widget
20160815 03:18:36< vultraz> if an image has a _grow key
20160815 03:18:43< vultraz> and it grows to fit available space
20160815 03:18:51< vultraz> it then tries to render the image in that space
20160815 03:19:36< vultraz> let's try this..
20160815 03:20:44< vultraz> oh for fucking hell
20160815 03:20:50-!- oldlaptop [~quassel@50-37-63-133.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20160815 03:24:15< tad_> Gee I love this stuff. "Find Baran. Baran sends you to kill leader for key. Take key to Baran." but what if you already found and killed the leader? Still tells you to find and kill him. So step back, step forward and you win.
20160815 03:25:02-!- Kwandulin [~Miranda@p200300760F35BFFF3C855A80BD25B667.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160815 03:25:28< vultraz> if I set the w/h and to image_width/height..
20160815 03:25:59< vultraz> that fixes the scaling
20160815 03:26:01< vultraz> but not
20160815 03:26:03< tad_> Yes. The aspect ratio should be sacrosanct.
20160815 03:26:15< vultraz> the intended use of width= height= in [image] :|
20160815 03:26:21< celmin> That's why I added ~SCALE_INTO for the help browser.
20160815 03:26:32< vultraz> I suppose i should just revert this feature
20160815 03:26:33-!- fabi_ [~fabi@176.4.53.88] has quit [Ping timeout: 240 seconds]
20160815 03:26:37< vultraz> and use ~SCALE()
20160815 03:26:37< celmin> Which feature?
20160815 03:26:45< vultraz> https://github.com/wesnoth/wesnoth/commit/05340bb81f661806fb6cc853be292738fabda1fe
20160815 03:27:13< celmin> What was the intent of adding it?
20160815 03:27:48< vultraz> was going to use it to make the campaign image area always a specific size
20160815 03:27:52< vultraz> in mp create
20160815 03:28:01< vultraz> so it didn't draw it at native size
20160815 03:28:07< vultraz> which I guess appending ~SCALE could do :|
20160815 03:29:14< vultraz> it's a bloody annoying issue
20160815 03:29:18< vultraz> this bug
20160815 03:29:24< vultraz> but I dunno how to fix it
20160815 03:31:19< vultraz> the image grows to fit available space, and the image is drawn at its w and h..
20160815 03:31:45< vultraz> i guess my problem was i wasn't sizing the image, i was sizing the widget
20160815 03:33:07< celmin> Hmm, so does it do something that ~SCALE() or ~SCALE_INTO() can't?
20160815 03:33:23< celmin> I don't really understand the motivation behind it, or that behind reverting it either.
20160815 03:33:33< vultraz> it does not
20160815 03:33:51< irker310> wesnoth: Charles Dang wesnoth:master 156ac4cf93d5 / / (4 files in 3 dirs): Revert "timage: allowed use of width/height keys for specific image size" https://github.com/wesnoth/wesnoth/commit/156ac4cf93d569fc3419fe918ec242f6c35965a8
20160815 03:34:01< vultraz> easiest solution
20160815 03:35:09< vultraz> tad_: should be fixed now
20160815 03:35:35< tad_> Grabbinb it now.
20160815 03:36:14< tad_> And, of course git pissed on a zillion files and it's gonna take 45 minutes to build. sigh
20160815 03:36:56< tad_> I don't get that. I checkout and the only changes are .cfg files for a campaign. And I checkout back to master and it says I changed EVERYTHING.
20160815 03:37:08< tad_> The timestamps are unchanged.
20160815 03:39:51< celmin> So I guess next step is PR730...
20160815 03:40:11< celmin> Oh, I need to get back to my PR too, there's an issue there that my testing didn't uncover.
20160815 03:41:30< tad_> I'm off looking at AToTB .. y'all take your time on the HttT PRs. I been over and over them and still find tweaks. So I'm stying away and only changing what is really in need of fixin'
20160815 03:52:46< tad_> What I liked was I noticed that Li'sar's troops went, "Villages? Wenches! We'll join the battle in a bit, m'am." and when I told 'em "No" .. OUCH .. they came at me hard! So .. let 'em have the wenches.
20160815 03:54:42-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160815 03:56:31-!- fabi_ [~fabi@176.0.125.63] has joined #wesnoth-dev
20160815 03:57:53< Aginor> ummm
20160815 03:58:04< Aginor> that's not a good look
20160815 04:01:23< celmin> ?
20160815 04:01:42< celmin> What's not?
20160815 04:08:40-!- hk238 [~kvirc@t224.ip7.netikka.fi] has joined #wesnoth-dev
20160815 04:12:03-!- Kwandulin [~Miranda@p200300760F35BFFF3C855A80BD25B667.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160815 04:17:27< vultraz> ?
20160815 04:20:40< celmin> Aginor?
20160815 04:22:12< vultraz> Aginor?
20160815 04:24:30-!- esr [~esr@wesnoth/developer/esr] has quit [Quit: WeeChat 1.3]
20160815 04:27:45-!- travis-ci [~travis-ci@ec2-54-198-237-197.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 04:27:46< travis-ci> wesnoth/wesnoth#10404 (master - 156ac4c : Charles Dang): The build has errored.
20160815 04:27:46< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152302892
20160815 04:27:46-!- travis-ci [~travis-ci@ec2-54-198-237-197.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 04:30:35< vultraz> blah
20160815 04:30:47< vultraz> have to mesh the configure and create engines in this dialog
20160815 04:31:28-!- esr [~esr@wesnoth/developer/esr] has joined #wesnoth-dev
20160815 04:33:39< vultraz> hm
20160815 04:33:45< vultraz> yeah, these weren't designed to mesh..
20160815 04:47:16< vultraz> i have two references
20160815 04:47:24< vultraz> to two different objects
20160815 04:47:57< vultraz> each of these takes a reference to a third object
20160815 04:49:37< vultraz> celmin: would any modifications by either a or b on c be reflected in the other?
20160815 04:50:04< celmin> Say what?
20160815 05:10:16-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20160815 05:20:27< vultraz> looks like i have to commit to the gui2 version
20160815 05:20:41< celticminstrel> ?
20160815 05:21:17< celticminstrel> If A and B both hold a reference to C, then any changes to C would be reflected both to A and B, by the way. If that's what you were asking.
20160815 05:21:39< vultraz> but if a changes c will b see the change too?
20160815 05:21:58< celticminstrel> I thought that was what I just said. If C is held by reference in both, then yes it will.
20160815 05:24:51< vultraz> celticminstrel: ok, perhaps you can help me with something
20160815 05:25:21< vultraz> not syntax, structure 
20160815 05:26:54< vultraz> so there are two very helpful classes
20160815 05:27:03< vultraz> configure_engine and create_engine
20160815 05:27:30< vultraz> as you can imagine, create is for the game selection screen, and configure is for the settings screen
20160815 05:27:53< vultraz> i need to make them work at the same time
20160815 05:28:52< vultraz> as far as I can tell, a saved_game reference is passed around 
20160815 05:30:09< vultraz> i need both create and configure engine objects in this dialog
20160815 05:30:40< vultraz> both should update the saved_game object, IIRC
20160815 05:31:56< vultraz> but the configure_engine expects the sides to already be set up when its construted..
20160815 05:32:25< Aginor> vultraz: are you trying to make this multi-threaded?
20160815 05:32:32< vultraz> what? no 
20160815 05:32:40< vultraz> I have no idea how to work with multithreading
20160815 05:32:51< vultraz> I'm combining two screens into one 
20160815 05:33:01< vultraz> for simplicity 
20160815 05:33:12< vultraz> (on the user's part, that is)
20160815 05:33:14< celticminstrel> Except apparently it's not as simple as it seems.
20160815 05:33:26< celticminstrel> Also, I still don't see how it can work on 800x600.
20160815 05:33:53< vultraz> are you advocating two dialogs? :|
20160815 05:34:09< celticminstrel> What the heck is wrong with Travis? It's not even loading the build results.
20160815 05:34:22< Aginor> break it apart into smaller work-objects that can be reordered
20160815 05:34:26< celticminstrel> Though it works when I reload...
20160815 05:34:34< vultraz> Aginor: ?
20160815 05:35:33< celticminstrel> So, the GCC 4.7 build bizarrely has all the WML tests fail.
20160815 05:35:38< Aginor> vultraz: break the game logic out of the dialogs, chuck it into a separate class
20160815 05:35:49< Aginor> or multiple
20160815 05:35:51< vultraz> Aginor: the logic is already split
20160815 05:36:05< vultraz> configure_engine and create_engine ARE seperate
20160815 05:36:11< celticminstrel> Okay, looks like the MP tests are the culprit.
20160815 05:36:16< Aginor> then, either call those classes on your game and have them act in the right order
20160815 05:36:49< vultraz> i guess I  need to move the code out of connect_engine's constructor..
20160815 05:36:52< vultraz> er
20160815 05:36:56< vultraz> configure
20160815 05:36:57< Aginor> OR call them from within the GUI handler (less good), or apply the strategy pattern so there's chance of reuse
20160815 05:37:03< celticminstrel> vultraz: So basically, the configure_engine cannot be constructed until the connect_engine is finalized, right?
20160815 05:37:33< vultraz> create_engine, not connect, but yes
20160815 05:37:42< celticminstrel> Yeah that.
20160815 05:38:14< celticminstrel> Then what you need to do is reconstruct the configure_engine if something changes in the create_engine, right?
20160815 05:38:15< vultraz> the configure_engine class expects the saved_game object to already have [side]s...
20160815 05:38:27< celticminstrel> The hacky way of doing it is manually calling the destructor and then using placement new... >_>
20160815 05:38:30< vultraz> no, both reference the same saved_game object
20160815 05:38:37< vultraz> I need some kind of...
20160815 05:38:45< vultraz> set_current_level function
20160815 05:38:55< vultraz> that exists for the create_engine..
20160815 05:39:11< celticminstrel> I don't understand how referencing the saved_game object makes the answer "no".
20160815 05:39:30< vultraz> I think I just need to give different data to the configure_engine every time you select a new game in the UI
20160815 05:42:52< vultraz> the point is, if i start editing one of the *_engine classes, i'll probably break the corresponding gui1 dialog
20160815 05:43:02< vultraz> meaning i need to commit to making this new gui2 dialog work
20160815 05:43:17< vultraz> you see?
20160815 05:43:18< celticminstrel> You could also re-split the dialog, though.
20160815 05:43:27< celticminstrel> Or do the hacky destroy-and-reconstruct trick.
20160815 05:43:38< vultraz> i don't want to split the dialog
20160815 05:44:00< celticminstrel> Does the configure_engine handle leader selection too?
20160815 05:44:12< vultraz> no I think that's the connect_engine
20160815 05:44:15< celticminstrel> I see.
20160815 05:44:36< celticminstrel> Aginor: Should I recommend the hacky destroy-and-reconstruct trick?
20160815 05:46:36< Aginor> celticminstrel: I don't know what you're referring to
20160815 05:46:51< celticminstrel> Aginor: x.~C(); new(&x) C()
20160815 05:47:25< Aginor> celticminstrel: no :D
20160815 05:49:14-!- TC02 [~quassel@venus.arosser.com] has quit [Ping timeout: 244 seconds]
20160815 05:50:56-!- TC02 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20160815 05:52:39< celticminstrel> vultraz: Is the configure engine required in order to build the connect_engine, or does that only need the saved_game?
20160815 05:53:21< vultraz> latter
20160815 05:54:14< vultraz> still trying to understand the code
20160815 05:55:07< celticminstrel> Okay, I think I'd recommend one of two strategies. 1) The dialog holds a shared_ptr<configure_engine> and builds a new one anytime something relevant changes. 2) Probably better, and possibly what you were thinking already - move the setup code from the configure_engine constructor to a function which the constructor then calls. Then the dialog holds an actual configure_engine (not a reference or pointer) and calls the update functi
20160815 05:55:07< celticminstrel> whenever something relevant changes.
20160815 05:57:59< Aginor> celticminstrel: observer-pattern? :D
20160815 05:58:34< celticminstrel> I don't think vultraz is skilled enough to pull that off, given that he gets confused by its use in places like the Lua console or inspector.
20160815 05:58:50< celticminstrel> (Was it used in the Lua console? I forget. Definitely in the inspector though.)
20160815 05:58:51< Aginor> hmm
20160815 05:59:02< Aginor> it'd make sense for something like this though
20160815 05:59:10< celticminstrel> Probably, yeah.
20160815 05:59:22< celticminstrel> Not entirely sure how it would solve his dilemma, mind you.
20160815 05:59:48< celticminstrel> (That is, the fact that the configure_engine relies on the create_engine being finalized.)
20160815 06:00:30< Aginor> no, that is a different kettle of fish, unfortunately
20160815 06:00:38< celticminstrel> Wait, did I get my patterns mixed up? The inspector uses model-view-controller. I suppose that's different from observer.
20160815 06:01:11< Aginor> but it seems to, at face value and without inspecting code, seems to be a fair assumption by the code
20160815 06:01:35< celticminstrel> I don't really understand the motivation for merging the two in the first place.
20160815 06:01:36< Aginor> celticminstrel: yes, observer pattern has the observed object keep a track of interested parties and notify them when it changes
20160815 06:01:47< celticminstrel> I see.
20160815 06:02:02< vultraz> I don't know anything about paradigms
20160815 06:02:06< vultraz> i just code what works
20160815 06:12:32-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160815 06:18:36< vultraz> what is this entry points business...
20160815 06:18:43< celticminstrel> ?
20160815 06:19:19< vultraz> https://github.com/wesnoth/wesnoth/blob/master/src/game_initialization/configure_engine.cpp#L28
20160815 06:20:20< vultraz> why is it pushing every scenario to a vector..
20160815 06:22:36< vultraz> you were saying something the other day about entry points
20160815 06:22:40< vultraz> vs SoS saves?
20160815 06:22:44< celticminstrel> Huh?
20160815 06:22:55< irker310> wesnoth: Celtic Minstrel wesnoth:master 07c228f74a69 / src/gui/dialogs/multiplayer/mp_create_game.cpp: MP Create: Minor fixups for OSX / XC4 https://github.com/wesnoth/wesnoth/commit/07c228f74a69d3a3552a93196166e70ce812316b
20160815 06:23:48< vultraz> could you just help me figure out what this code does :|
20160815 06:23:53< celticminstrel> I guess "entry points" means "initial scenarios".
20160815 06:24:05< celticminstrel> So, for example, in an MP campaign, only the first scenario qualifies.
20160815 06:26:22< vultraz> hm
20160815 06:26:30< vultraz> ok, putting that aside for now
20160815 06:26:37< vultraz> if I'm reading the code correctly
20160815 06:26:46< vultraz> the side_cfg_ variable holds the data
20160815 06:26:48< vultraz> right?
20160815 06:28:10< celticminstrel> I guess?
20160815 06:28:23< celticminstrel> It looks like it only holds side 1.
20160815 06:28:23< vultraz> so I need to update it
20160815 06:29:42< vultraz> why is child_or_empty const..
20160815 06:29:43< celticminstrel> It shouldn't be writing to std::cerr. :/
20160815 06:30:00< vultraz> that bit needs to be removed
20160815 06:30:03< vultraz> anyway
20160815 06:30:03< celticminstrel> Because if the child doesn't exist, it returns an empty static config.
20160815 06:30:23< vultraz> ok, i'll just use child()
20160815 06:30:26< celticminstrel> Making it non-const would make it possible that it be modified, meaning later calls won't actually return an empty config.
20160815 06:30:45< vultraz> wait, no, that doesn't work
20160815 06:30:47< vultraz> UGH
20160815 06:30:55< vultraz> i forgot child() can't handle missing wml nodes
20160815 06:30:58< vultraz> what to dooo
20160815 06:31:00< celticminstrel> You can use child(), but then you need to check .valid().
20160815 06:31:14< celticminstrel> I don't think that's necessarily the right approach.
20160815 06:31:31< vultraz> I can't reassign side_cfg_ if it's cost
20160815 06:31:32< celticminstrel> (The existence of .valid() is kind of weird to begin with.)
20160815 06:31:34< vultraz> const
20160815 06:31:43< vultraz> so i need to make it a non-const reference..
20160815 06:31:47< vultraz> unless I make it a copy?
20160815 06:31:50< celticminstrel> Make it a copy.
20160815 06:31:51< vultraz> yes, probably should make it a copy..
20160815 06:32:22< celticminstrel> Or a pointer.
20160815 06:32:27< vultraz> *gasp*
20160815 06:32:33< celticminstrel> Copy is probably easier to deal with, I guess.
20160815 06:33:20< celticminstrel> Still, as a pointer, it'd be a pointer-to-const, so you could still use child_or_empty and avoid unnecessary copies, but also update it when needed.
20160815 06:33:47< celticminstrel> Actually, there's not really any need to ever update it unless there are scenarios that have no sides, which is a problem in itself.
20160815 06:33:48< vultraz> i'll look into that later
20160815 06:33:53< celticminstrel> (In the current model, I mean.)
20160815 06:34:05< vultraz> celticminstrel: a scenario will have no sides when the dialog is opened
20160815 06:34:06< celticminstrel> Wait, no, that's probably not true.
20160815 06:34:10< celticminstrel> Never mind.
20160815 06:34:13< vultraz> no scenario is selected until the end of pre_show
20160815 06:34:20< vultraz> pre_show is not the constructor
20160815 06:34:25< celticminstrel> Right.
20160815 06:34:48< vultraz> that's what im trying to work around here
20160815 06:34:53< vultraz> that period where it's empty
20160815 06:34:57< celticminstrel> So there's a major problem here that can't be resolved if you want tmp_create to contain an object of type configure_engine.
20160815 06:35:06< celticminstrel> It's line 25.
20160815 06:35:20< vultraz> i removed that
20160815 06:35:24< celticminstrel> That's not good.
20160815 06:35:30< vultraz> :|
20160815 06:35:33< celticminstrel> Because having no sides is, in fact, an error.
20160815 06:35:43< celticminstrel> A scenario cannot function with no sides.
20160815 06:35:48< vultraz> that should be moved to connect_engine, then
20160815 06:36:01< vultraz> or somewhere..
20160815 06:36:20< celticminstrel> I assume it's in configure_engine because configure_engine needs to manipulate sides.
20160815 06:36:23< celticminstrel> Am I wrong?
20160815 06:36:46< celticminstrel> If so, then yeah, it could be moved to connect_engine.
20160815 06:37:10< vultraz> configure engine handles side-related stuff, yes
20160815 06:37:18< celticminstrel> ...I guess the fact that side_cfg_ is const suggests that I am, in fact, wrong.
20160815 06:38:09< celticminstrel> Using a pointer is probably a good choice actually, if you want to avoid the copy; but using a copy is probably fine as well.
20160815 06:38:26< celticminstrel> I mean, this is not one of the areas of the game where performance is a significant issue.
20160815 06:40:02< vultraz> hm
20160815 06:40:04< vultraz> wait
20160815 06:40:13< vultraz> the parameters for the scenario settings come from..
20160815 06:40:19< vultraz> a mp_game_settings reference
20160815 06:40:26< vultraz> the fuck is this
20160815 06:41:52< vultraz> a struct
20160815 06:42:27< vultraz> guess I need to update that too
20160815 06:42:59< vultraz> wesnoth has so much misdirection
20160815 06:43:21< vultraz> hm
20160815 06:43:26< vultraz> it's initialized from state_..
20160815 06:43:37< vultraz> does the create engine touch state?
20160815 06:44:18< vultraz> prepare_for_new_level
20160815 06:44:20< vultraz> let's try that
20160815 06:44:49< vultraz> nope
20160815 06:46:18< vultraz> god dammit
20160815 06:46:46< vultraz> how does this not work
20160815 06:46:58< vultraz> it calls a function to update state_.mp_settings()
20160815 06:47:15< vultraz> then another function is called in the other engine that sets a member to state_.mp_settings()
20160815 06:47:17< vultraz> :|
20160815 06:47:47< vultraz> this should logically work!
20160815 06:55:04-!- JyrkiVesterinen [~jyrki@87-92-6-34.bb.dnainternet.fi] has joined #wesnoth-dev
20160815 06:55:52-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160815 06:55:52-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160815 06:57:05< vultraz> i can't understand it
20160815 07:02:05< vultraz> maybe it's this get_starting_pos thing
20160815 07:02:27< vultraz> hm.. no.. it shouldn't..
20160815 07:05:07< vultraz> the value of paramater_'s turns is 0
20160815 07:05:10< vultraz> parameter 
20160815 07:05:12< vultraz> why
20160815 07:07:23< vultraz> wait
20160815 07:07:29< vultraz> it's not writing to mp_settings
20160815 07:07:33< vultraz> then what the *hell* is
20160815 07:08:20< vultraz> ok, it sets some parameters
20160815 07:08:22< vultraz> but..
20160815 07:08:33< vultraz> still like turns is initiated elsewhere
20160815 07:08:34< vultraz> but where
20160815 07:10:51< vultraz> ok, preferences::level has nothing to do with this..
20160815 07:12:08< vultraz> maybe gfgtdf can help me
20160815 07:12:18< vultraz> he worked with this code, I think
20160815 07:12:21< vultraz> or perhaps iceiceice
20160815 07:16:16-!- travis-ci [~travis-ci@ec2-54-198-237-197.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 07:16:17< travis-ci> wesnoth/wesnoth#10405 (master - 07c228f : Celtic Minstrel): The build has errored.
20160815 07:16:17< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152318320
20160815 07:16:17-!- travis-ci [~travis-ci@ec2-54-198-237-197.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 07:33:19-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160815 07:35:40-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds]
20160815 07:35:40-!- wedge010 is now known as wedge009
20160815 07:40:09-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160815 07:45:42-!- celticminstrel is now known as celmin|sleep
20160815 07:47:02-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Ping timeout: 244 seconds]
20160815 07:57:46-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160815 08:04:42-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Ping timeout: 258 seconds]
20160815 08:05:38-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160815 08:15:09-!- iwaim__ [~iwaim@2001:2c0:40e:2002:0:4:14:80] has quit [Ping timeout: 260 seconds]
20160815 08:46:12-!- iwaim__ [~iwaim@2001:2c0:40e:2002:0:4:14:80] has joined #wesnoth-dev
20160815 08:57:42< vultraz> 	/// @return the config from which the game will be started. (this is [scenario] or [snapshot] in the savefile)
20160815 08:57:43< vultraz> 	config& get_starting_pos();
20160815 08:57:47< vultraz> what kind of name for a function is this :|
20160815 08:57:56< vultraz> IT HAS NOTHING TO DO WITH STARTING POSITIONS
20160815 09:02:40< DeFender1031> vultraz, it's the starting position of EVERYTHING. :P
20160815 09:03:05< vultraz> :P
20160815 09:03:10< DeFender1031> but yes, "starting position" already means something else in wesnoth terminology, so that's a terrible name.
20160815 09:03:40< vultraz> i shall change it
20160815 09:03:50< vultraz> along with that get_shared_ptr function which doesn't get a shared ptr :P
20160815 09:05:25< irker310> wesnoth: ln-zookeeper wesnoth:master d2f528eec92b / data/core/macros/ai_controller.cfg: Prevent extraneous "="'s from appearing in some AI controller menus https://github.com/wesnoth/wesnoth/commit/d2f528eec92bf0e99a27b42e5c1ab989c5f4f987
20160815 09:10:55< irker310> wesnoth: ln-zookeeper wesnoth:1.12 0dd2e9658492 / data/core/macros/ai_controller.cfg: Prevent extraneous "="'s from appearing in some AI controller menus https://github.com/wesnoth/wesnoth/commit/0dd2e965849243080d6874de38079ef424a527ce
20160815 09:16:48< vultraz> gfgtdf: do you know where in the mp game initialization state.mp_settings() gets updated?
20160815 09:37:48< Aginor> vultraz: design patterns is about common approaches to common problems, so that you don't have to go and figure out what goes on every time, you look at it and go "oh yeah, they did that"
20160815 09:38:35< Aginor> (I'm continuing on from ~4h ago)
20160815 09:48:43 * zookeeper sees let's play where person thinks they have to issue the AI controller orders for each allied _unit_ individually, facepalms
20160815 09:49:27< zookeeper> far from the only facepalm moment, but still that was the most painful one
20160815 09:50:28< zookeeper> https://youtu.be/QpiyX2EeC5w?t=2147
20160815 09:51:14< zookeeper> well, luckily it seems like he figured it out
20160815 09:59:40-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 09:59:41< travis-ci> wesnoth/wesnoth#10406 (master - d2f528e : ln-zookeeper): The build has errored.
20160815 09:59:41< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152342079
20160815 09:59:41-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 10:12:50< vultraz> Aginor: that doesn't help if you don't understand the pattern :)
20160815 10:13:06< vultraz> or in cases where the pattern just makes the code worse
20160815 10:13:34< vultraz> (ie, m-v-c in the lua console)
20160815 10:15:17< Aginor> vultraz: reading opportunity ;)
20160815 10:16:22< vultraz> however, my problem right now is I can't find a specific piece of code
20160815 10:17:09< Aginor> ok
20160815 10:17:09< vultraz> I wish i had an IDE with the ability to provide me the trail of factors that affects a specific object 
20160815 10:17:33< Aginor> try eclipse
20160815 10:17:36< Aginor> it's pretty good
20160815 10:22:39< Aginor> bah
20160815 10:22:46< Aginor> slither.io is too laggy today
20160815 10:33:57-!- JyrkiVesterinen [~jyrki@87-92-6-34.bb.dnainternet.fi] has quit [Quit: .]
20160815 11:08:41-!- fabi__ [~fabi@176.5.101.42] has joined #wesnoth-dev
20160815 11:08:41-!- fabi_ [~fabi@176.0.125.63] has quit [Read error: Connection reset by peer]
20160815 11:22:17-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160815 11:36:08-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has joined #wesnoth-dev
20160815 11:36:50< gfgtdf> 20160815 09:16:48< vultraz> gfgtdf: do you know where in the mp game initialization state.mp_settings() gets updated?
20160815 11:37:33< gfgtdf> vultraz: in the configure_engine::set_.. function see here https://github.com/wesnoth/wesnoth/blob/master/src/game_initialization/configure_engine.cpp#L86
20160815 11:38:35< gfgtdf> vultraz: then the values are wirtten sin the differenet side here in configure_engine::write_parameters  https://github.com/wesnoth/wesnoth/blob/master/src/game_initialization/configure_engine.cpp#L208
20160815 11:38:57< vultraz> ohhhhhhhhh
20160815 11:39:08-!- fabi_ [~fabi@176.0.18.39] has joined #wesnoth-dev
20160815 11:40:31< gfgtdf> vultraz: this ins the function inth gui1 dialog that calls all there https://github.com/wesnoth/wesnoth/blob/36d14951c68d370ae3d8df391c72d2ab441fa534/src/game_initialization/multiplayer_configure.cpp#L291
20160815 11:42:03-!- fabi__ [~fabi@176.5.101.42] has quit [Ping timeout: 258 seconds]
20160815 11:42:19< vultraz> hmmmmmm
20160815 11:43:23< gfgtdf> vultraz: i actuall think you can just reuse the configure_engine object.
20160815 11:43:24< vultraz> ok, that's what's sets them
20160815 11:43:32< vultraz> gfgtdf: yes im trying to do that
20160815 11:43:37< vultraz> but im doing this in one dialog
20160815 11:43:41< vultraz> instead of two
20160815 11:44:21< vultraz> i just need to find how to get the initial value of certain sliders from the config...
20160815 11:44:23< vultraz> like turns=
20160815 11:44:55< gfgtdf> vultraz: for that there are also function ins configure_engine: https://github.com/wesnoth/wesnoth/blob/master/src/game_initialization/configure_engine.cpp#L129
20160815 11:45:14< vultraz> ohh
20160815 11:45:16< vultraz> thatt
20160815 11:46:13< vultraz> ok
20160815 11:46:26< vultraz> I should just need a very minor change to the configure_engine then
20160815 12:08:32-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160815 12:10:57-!- irker310 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160815 12:13:26 * celmin|sleep doesn't think m-v-c makes the code worse in the inspector (in the Lua console it's debatable).
20160815 12:14:05< celmin|sleep> (I kinda think it may be overengineering in the Lua console, since there's only one possible view, but whatever. Probably not to the point of being worth changing.)
20160815 12:20:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160815 12:28:05-!- Kwandulin [~Miranda@2003:76:f35:bf31:61cc:c0a2:c5d0:6e47] has joined #wesnoth-dev
20160815 12:38:11< vultraz> ok, so now I need to figure out how to update the scenario returned by get_starting_pos
20160815 12:41:16< vultraz> it fetches it from state...
20160815 12:41:22< vultraz> do I need to modify state?
20160815 12:42:14< vultraz> wait, wait
20160815 12:44:13< vultraz> then again, since it's referencing the same object it shouldn't matter..
20160815 12:44:39< vultraz> yeah
20160815 12:44:41< vultraz> fuck
20160815 12:48:18< vultraz> gfgtdf: ok, so, I'm trying to make configure_engine keep up with the values of each scenario as you select them in the dialog. SO I added a function that did this: side_cfg_ = state_.get_starting_pos().child_or_empty("side"); But it only seems to show the data for the first scenario no matter what you click. If I understand correctly, that's because get_starting_pos keeps pointing to the...
20160815 12:48:20< vultraz> ...same thing. do you know any way i can make that point to a new game's data?
20160815 12:50:10< vultraz> tho...
20160815 12:50:11< vultraz> hm
20160815 12:50:21< vultraz> I guess that actually means swapping the whole state_
20160815 12:50:23< vultraz> but..
20160815 12:50:33< vultraz> both engine objects operate on the same state
20160815 12:50:39< vultraz> :|
20160815 12:52:56< vultraz> obviously in the gui1 version the state is edited somehow..
20160815 12:55:56-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160815 12:56:14< vultraz> even in gui1 they're working on the same object
20160815 12:56:19< vultraz> so what. changes it!
20160815 12:57:21< vultraz> one would think ... expand_scenario()?
20160815 12:57:41< vultraz> which is in create_engine::prepare_for_new_level()
20160815 12:57:46< celmin|sleep> I don't really even know what you're talking about, so it's hard to help. (I haven't looked much at the code.)
20160815 12:57:50-!- celmin|sleep is now known as celticminstrel
20160815 12:58:40< vultraz> it's very roundabout
20160815 12:58:49< celticminstrel> So the server seems to have a crash again.
20160815 12:59:01< celticminstrel> Not sure if it's the same one, haven't debugged yet.
20160815 12:59:34< vultraz> i call both ...
20160815 12:59:37< vultraz> 	create_engine_.prepare_for_scenario();
20160815 12:59:37< celticminstrel> This crash is responsible for the Travis tests erroring.
20160815 12:59:38< vultraz> 	create_engine_.prepare_for_new_level();
20160815 12:59:50< vultraz> if that edits state_..
20160815 13:00:12< vultraz> since it's a reference the config_engine should then return a new side in get_starting_pos()
20160815 13:00:20< vultraz> since saved_game::expand_scenario assigns it
20160815 13:01:15< vultraz> prepare_for_new_level even seems to be there before the dialog closes in gui1..
20160815 13:01:33< celticminstrel> Unrelatedly, it's almost impossible for me to follow the log, since the three processes are all jumbled up.
20160815 13:01:58< vultraz> so side_cfg_ = state_.get_starting_pos().child_or_empty("side"); should work!
20160815 13:02:00< vultraz> but it doesn't..
20160815 13:02:13< celticminstrel> ...eh? Is that a WML error? :|
20160815 13:03:02-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160815 13:03:21< celticminstrel> The crash appears to be the same as before.
20160815 13:03:28< celticminstrel> That is, invalid gzip data.
20160815 13:03:50< gfgtdf> vultraz: usually do you change state_
20160815 13:04:18< vultraz> ?
20160815 13:04:35< gfgtdf> vultraz: the create_engine::prepare function usually oberwrite the state_ object cmpleteley
20160815 13:05:23< gfgtdf> .expand_Sceario reads a [scenario] from the game confi into the state_ object.
20160815 13:05:27-!- boucman_work [~boucman@2a02-8428-034f-f800-9e32-0c7c-b391-6223.rev.sfr.net] has quit [Quit: Leaving.]
20160815 13:05:55< vultraz> right
20160815 13:06:31< vultraz> this is the order I do stuff http://pastebin.com/uXpkqXVG
20160815 13:06:33-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20160815 13:06:34-!- wedge010 is now known as wedge009
20160815 13:07:18< vultraz> configure_engine::update_side_cfg just updates its side_cfg_ member.
20160815 13:07:48< vultraz> so logically, next time I call configure_engine::*_default(), it should give me the new value..
20160815 13:07:50< vultraz> but it doesn't
20160815 13:09:28< gfgtdf> vultraz: hmm side_cfg_  is a reference, so i wonder who you can update it ?
20160815 13:11:02< vultraz> i made it a copy
20160815 13:11:23< vultraz> and non-const
20160815 13:11:55< gfgtdf> vultraz: if you want to stick closer to the gui1 code, you can also have something like unique_ptr<connect_engine> connect_engine_; and then just do connect_engine.reset(new connect_engine(state_)) ehnever you select a different scneairo.
20160815 13:12:47< vultraz> i believe celmin suggested that..
20160815 13:12:57< vultraz> but really i don't care about sticking to the gui1 method
20160815 13:13:02< vultraz> (also you mean configure_engine?)
20160815 13:13:17< gfgtdf> vultraz: yes i meant confugur_engine
20160815 13:13:22< celmin> I did suggest that (though I said shared_ptr, but unique_ptr also works, maybe better).
20160815 13:20:25< celmin> I think that method would probably be easiest though.
20160815 13:21:45< celmin> I wonder if it's possible to write a Lua plugin to do load-testing on the MP server.
20160815 13:22:17< celmin> I seem to recall performance was one of the complaints of the GUI2 lobby?
20160815 13:23:19< vultraz> that, and horrible listbox behavior
20160815 13:23:23< vultraz> like shrinking rows
20160815 13:23:33< vultraz> we need to fix that
20160815 13:27:48-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160815 14:00:50< vultraz> whyy are these sliders not moving
20160815 14:03:14< vultraz> even using the unique_ptr method :|
20160815 14:11:49-!- irker932 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160815 14:11:49< irker932> wesnoth: Charles Dang wesnoth:master e31615b68771 / / (4 files in 3 dirs): MP Create: bunch of WIP stuff https://github.com/wesnoth/wesnoth/commit/e31615b687710c5e281c0c10f148f1524ebe82a6
20160815 14:11:58< vultraz> i really have no idea why it doesn't work
20160815 14:12:14< vultraz> celmin: maybe you could give the code a look
20160815 14:12:20< vultraz> I'm absolutely stumped
20160815 14:12:33< celmin> Sure, I'll take a look.
20160815 14:13:41< irker932> wesnoth: Charles Dang wesnoth:master 168cc67e04e7 / src/game_initialization/ (configure_engine.cpp configure_engine.hpp): ng::configure_engine: enable setting of side_cfg_ https://github.com/wesnoth/wesnoth/commit/168cc67e04e7a271390631d349e7d0c787661c34
20160815 14:15:25< celmin> "bunch of WIP stuff" is a terrible, terrible commit message.
20160815 14:16:44< vultraz> granted
20160815 14:17:10< celmin> Is it my imagination or are you now showing the GUI1 MP Create after the GUI2 one closes?
20160815 14:17:59< vultraz> yeah
20160815 14:18:11< celmin> I still don't understand the motivation of merging create and configure.
20160815 14:18:27< celmin> And looking at it just now, I think there's a fairly strong reason not to.
20160815 14:18:36< vultraz> for one, it was supposed to make it easier for players to see the settings of the map they're choosing
20160815 14:18:52< vultraz> without flipping back and forth between screens
20160815 14:19:03< celmin> Which settings?
20160815 14:19:31< vultraz> fog, shroud, village gold, etc
20160815 14:19:33< vultraz> exp modifiers
20160815 14:19:37< vultraz> all that stuff
20160815 14:19:48< celmin> You could always just put a preview of those in the create screen, rather than merging the two screens.
20160815 14:19:59< celmin> I'll state the reason I saw, though.
20160815 14:20:07< vultraz> what is this reason
20160815 14:20:12< vultraz> (and don't say 800x600 :| )
20160815 14:20:20< celmin> Well yes, that's another reason. :P
20160815 14:20:33< celmin> I could be wrong since I don't really understand this code, but..
20160815 14:21:10< celmin> Around line 553 in g_i/multiplayer.cpp
20160815 14:21:42< celmin> It appears that if the player selects Load Game, the configure stage is skipped - it goes straight to the "connect" stage.
20160815 14:22:23< vultraz> yes
20160815 14:22:32< vultraz> i haven't decided how to handle load game yet
20160815 14:22:49< celmin> There's also a third reason, but it's kinda weaker - single-player may require a configuration stage if modifications are enabled. (I say it's weaker because it may be better to have a separate dialog for it.)
20160815 14:23:09< vultraz> and?
20160815 14:23:25< vultraz> mods are handled here too
20160815 14:23:31< celmin> Why are you saying "and?"?
20160815 14:23:53< vultraz> because it's irrelevant if sp needs configuration or not
20160815 14:24:16< celmin> I did say it's weaker, because it might be better to have a separate SP config dialog.
20160815 14:25:20< celmin> Anyway, load game seems like a somewhat strong reason to keep them separate.
20160815 14:25:21< vultraz> only thing unique about sp is the ()*@#)( entry point stuff.
20160815 14:25:36< celmin> Oh right, I forgot to finish looking over your code.
20160815 14:25:51< vultraz> i could always just proceed to the next dialog if one loads game
20160815 14:25:57< vultraz> there's no reason i couldn't
20160815 14:26:21< celmin> I suppose. I'm still questioning the merit of merging them though.
20160815 14:27:03< celmin> As I said, if what you want is for the player to be able to see the preset map settings, there are other ways to do it.
20160815 14:27:47< vultraz> also, I feel it's better to utilize the available space and keep the game setup process in one place
20160815 14:28:00< celmin> The available space is an illusion, though.
20160815 14:28:18< vultraz> disregarding that
20160815 14:28:29< celmin> Ignoring it doesn't make it go away.
20160815 14:28:29< vultraz> it's better to keep this stuff in one place
20160815 14:28:37< celmin> Why?
20160815 14:28:43< vultraz> it just is
20160815 14:28:59< celmin> So you're just doing it on a whim. That's not a very good reason.
20160815 14:29:02< vultraz> it keeps info available and in one place
20160815 14:29:17< vultraz> it allows selection of different games without switching back and forth
20160815 14:30:18< vultraz> it allows one to quickly see different modifications for different selections of mods/eras without toggling back and forth
20160815 14:30:27< vultraz> you still need to toggle a tab, granted
20160815 14:32:28< vultraz> and again, it's merging two screens!
20160815 14:32:47< vultraz> how is reducing the number of screens a player needs to go through to create a game bad
20160815 14:32:50< celmin> Merging two screens is not a valid argument.
20160815 14:33:04< celmin> There was probably a reason for them to be separate in the first place, too.
20160815 14:33:13< vultraz> Maybe?
20160815 14:33:18< vultraz> But who knows at this point
20160815 14:33:25< celmin> And going through one extra screen honestly isn't that much of a pain in my opinoin.
20160815 14:33:27< vultraz> That's ancient history 
20160815 14:33:29< celmin> ^opinion
20160815 14:33:41< vultraz> Why is this such a point of contention with you :|
20160815 14:33:45< celmin> Just because it's ancient history doesn't mean it's irrelevant. (It may be, or may not be.)
20160815 14:35:27< vultraz> let's see what zookeeper thinks
20160815 14:35:47< vultraz> zookeeper: do you support mp create and configure becoming one screen?
20160815 14:35:56< vultraz> or at least, would you not object :P
20160815 14:36:01< celmin> It's not that I hate the idea of merging the two screens, but I think you're just not thinking it through clearly and considering all the advantages, disadvantages, consequences, whatever.
20160815 14:36:21< vultraz> the only disadvantages are for us, the people implementing it
20160815 14:36:29< vultraz> there's 0 disadvantage for the player
20160815 14:36:34< vultraz> that's reason enough
20160815 14:36:47< zookeeper> vultraz, in principle of course it's a good idea. whether you can cram everything into one screen in practise is another thing.
20160815 14:36:52< celmin> No, I've already given two disadvantages that relate to the player.
20160815 14:37:09< celmin> So your saying that is just proof that you're not thinking it through.
20160815 14:37:26< vultraz> 800x600 and...?
20160815 14:37:40< vultraz> Load Game behavior?
20160815 14:37:43< vultraz> both have solutions
20160815 14:37:52< celmin> Well, admittedly, the other one affects the GUI1 version too.
20160815 14:38:06< zookeeper> what are the disadvantages?
20160815 14:38:26< celmin> The main disadvantage for me is that I don't know it it can fit in 800x60.
20160815 14:38:28< celmin> ^600
20160815 14:38:56< vultraz> look
20160815 14:39:14< vultraz> I can make the details area another tab at that resolution if needs be
20160815 14:39:15< celmin> There might be other disadvantages I haven't thought of. I dunno.
20160815 14:39:25< vultraz> you've demonstrated different layouts are possible
20160815 14:39:33< celmin> Yeah, I have.
20160815 14:40:06< celmin> Not entirely sure if that would allow something that's not a tab in one layout to become a tab in another layout. (It might be possible with careful programming?)
20160815 14:40:55< vultraz> we shall see
20160815 14:41:51< zookeeper> re-paste the latest screenshot?
20160815 14:42:39< celmin> vultraz: Also, maybe multiple screenshots this time?
20160815 14:42:48< celmin> vultraz: Since there are multiple tabs.
20160815 14:43:00< vultraz> zookeeper: https://drive.google.com/file/d/0B-mR9s8FduLLSHd2U3JTRWRGVkU/view?usp=sharing , https://drive.google.com/file/d/0B-mR9s8FduLLQlVzcTM0X0FQRTA/view?usp=sharing , and https://drive.google.com/file/d/0B-mR9s8FduLLMGh0QmRNZUVCbFk/view?usp=sharing
20160815 14:43:03< vultraz> very slightly out of date
20160815 14:43:13< vultraz> but that's the gist of it
20160815 14:45:32< zookeeper> one thing you might want to consider is that the player won't automatically be taken through the settings screen anymore.
20160815 14:46:00< vultraz> true
20160815 14:46:03< celmin> That's a good point … since it's a separate tab, they could easily just skip it.
20160815 14:46:11< vultraz> but one also has to take into account that a player might not want to
20160815 14:46:17< celmin> Though, I'm not sure it's a problem.
20160815 14:46:49< zookeeper> so you might want to add some kind of extra hint or confirmation or something to make sure they know that they probably want to go check out the settings
20160815 14:47:23< vultraz> sure
20160815 14:47:35< vultraz> anyway, im going to get some sleep
20160815 14:47:48< zookeeper> of course usually the defaults are all fine and good, but IIRC the game remembers at least some of that stuff, in which case it might not be obvious that when you create a new game it might use the same settings as you used in a previous one
20160815 14:47:54< vultraz> ill read feedback in the morning
20160815 14:48:35< celmin> Yeah, it does remember your settings.
20160815 14:54:33-!- edgrey [~edgrey@178.205.2.70] has joined #wesnoth-dev
20160815 14:58:23-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160815 14:59:56< tad_> voltraz: I finally got around to looking at the recruit/recall/create panel images. The still look bad, to me. I have an example in my fork issues with a reference scaled-up 144x144 Bowman for comparison
20160815 15:06:00-!- travis-ci [~travis-ci@ec2-54-198-237-197.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 15:06:01< travis-ci> wesnoth/wesnoth#10408 (master - e31615b : Charles Dang): The build has errored.
20160815 15:06:01< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152396274
20160815 15:06:01-!- travis-ci [~travis-ci@ec2-54-198-237-197.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 15:20:52-!- oldlaptop [~quassel@50-37-63-133.mskg.mi.frontiernet.net] has quit [Remote host closed the connection]
20160815 15:22:05-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160815 15:26:50-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160815 15:28:57-!- Bonobo [~Bonobo@2001:44b8:254:3200:407:dc4d:a335:7099] has quit [Quit: Leaving]
20160815 15:49:11-!- midzer [~quassel@p5B296F88.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160815 15:49:18< irker932> wesnoth: ln-zookeeper wesnoth:master 471e92cd181b / data/campaigns/Liberty/scenarios/05_Hide_and_Seek.cfg: Added a reminder note about how vision range is determined https://github.com/wesnoth/wesnoth/commit/471e92cd181b0ca6e45086fbde3bcc5534a59005
20160815 15:55:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 16:04:05-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has joined #wesnoth-dev
20160815 16:06:22< gfgtdf> 20160815 14:25:21< vultraz> only thing unique about sp is the ()*@#)( entry point stuff.
20160815 16:06:30< gfgtdf> vultraz: the entry pint stuff is afaik never used in sp.
20160815 16:07:54< irker932> wesnoth: ln-zookeeper wesnoth:master 9eb4c289a493 / data/campaigns/Eastern_Invasion/scenarios/07a_The_Crossing.cfg: EI S7a: changed Ancient Lich back to the plain Lich it should be https://github.com/wesnoth/wesnoth/commit/9eb4c289a493c2ee68189d3a3e51674803e51b53
20160815 16:10:58< gfgtdf> vultraz, celmin any opinions on changing that 'if (gui2::new_widgets)' in multipler.cpp to if 'preferences::new_lobby()' so that using the new lobby in preferences wodul automtically use the new create screen too ?
20160815 16:11:24< gfgtdf> vultraz, celmin: the advantage it that its easier to test the new create screen without rebuiling wesnoth
20160815 16:11:37< celmin> That would be great, though it could possibly wait until the new MP Create is actually functional.
20160815 16:12:43< gfgtdf> celmin: hmm i guess wuif its not functionail until the next releae we can just rever tif before that release
20160815 16:12:55< celmin> I suppose.
20160815 16:15:39< irker932> wesnoth: ln-zookeeper wesnoth:master dbcc99a421e1 / data/campaigns/Eastern_Invasion/scenarios/07b_Undead_Crossing.cfg: EI S7b: removed unnecessary and annoying [delay]s https://github.com/wesnoth/wesnoth/commit/dbcc99a421e1f319f1ab497f0f5ee2e636f7d6f8
20160815 16:17:33< irker932> wesnoth: ln-zookeeper wesnoth:master a3f1d199a545 / data/campaigns/Liberty/scenarios/08_Glory.cfg: L S8: keep view centered on Halstead a bit longer during the intro https://github.com/wesnoth/wesnoth/commit/a3f1d199a545a07bccc58eb7cdb2e8e4d21ef432
20160815 16:18:55< gfgtdf> any opinion adaing to rmeove  wesnoth.fire_event_by_name (whcih si an alias for wesnoth.fire_event ) ?
20160815 16:19:48-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has joined #wesnoth-dev
20160815 16:19:51< celmin> Hmm, I don't really mind.
20160815 16:20:25< celmin> I'd prefer if there was a way for fire_event to handle both cases, but I don't think there's any good way to do that (a boolean parameter is not a good way).
20160815 16:26:49-!- fabi_ [~fabi@176.0.18.39] has quit [Ping timeout: 244 seconds]
20160815 16:33:39-!- Kwandulin [~Miranda@2003:76:f35:bf31:61cc:c0a2:c5d0:6e47] has quit [Read error: Connection reset by peer]
20160815 16:50:01< gfgtdf> vultraz whem im trying the experimental mp create i get 'no side found sneario aborting'
20160815 17:10:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 17:16:38-!- ancestral [~ancestral@67-4-224-82.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160815 17:23:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 17:25:37-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160815 17:49:40-!- hk238 [~kvirc@t224.ip7.netikka.fi] has quit [Quit: http://www.kvirc.net/ 4.9.1 Aria]
20160815 17:56:41-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 17:56:43< travis-ci> wesnoth/wesnoth#10412 (master - dbcc99a : ln-zookeeper): The build has errored.
20160815 17:56:43< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152426141
20160815 17:56:43-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 18:34:00-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has joined #wesnoth-dev
20160815 18:34:01< travis-ci> wesnoth/wesnoth#10413 (master - a3f1d19 : ln-zookeeper): The build has errored.
20160815 18:34:01< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/152426638
20160815 18:34:01-!- travis-ci [~travis-ci@ec2-54-145-118-127.compute-1.amazonaws.com] has left #wesnoth-dev []
20160815 18:58:59-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160815 19:00:29-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 19:02:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160815 19:07:33-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has joined #wesnoth-dev
20160815 19:11:02< gfgtdf> vultraz: on thing that bothers me in the new gui2 mpcreate is that, in order to show the 'settings' you need to load the scenario config, and to load the scneairo config you need might need to reload the cache if that scneairo uses custom era_define= or campaign_define.
20160815 19:11:44< gfgtdf> vultraz: now since teh settings are showy whenever you select a different map in that listbox it migth happen that when you click ona  listbox item the game config is relaoded and a loadingsceen appears.
20160815 19:15:28-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 19:17:39-!- irker932 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160815 19:22:35< gfgtdf> vultraz: this is even more true in the 'campaigns' section where most of them have campaign_defines
20160815 19:24:09< celmin> I thought campaign_defines was not multiplayer-friendly.
20160815 19:24:55< celmin> I distinctly recall aeth trying to use it to make wolves advance, and getting out-of-sync errors. Unless that was fixed at some point?
20160815 19:25:41< gfgtdf> celmin: hmm i don't realyl see how thats rleated to camapign_defines.
20160815 19:26:04< celmin> How is it not related to campaign_defines?
20160815 19:26:17< celmin> Wolf advancement is enabled by a campaign_define.
20160815 19:27:59< gfgtdf> celmin: well you cna use campaign_defines liek that, even in mp. Still i dont liek that suhc advancements are usualyl enabled by defines, i mean you have to rleoad the whoile game config just to chagne oen advancment...)
20160815 19:28:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160815 19:28:13< gfgtdf> celmin: LoW uses campaign_defines also in 1.12
20160815 19:28:35< gfgtdf> celmin: you cna see that it reloads the game_config (loadingscreen) between the mp_create adn teh mp_configure screen
20160815 19:28:38< gfgtdf> can*
20160815 19:28:48< celmin> What campaign_defines does LoW use?
20160815 19:30:10< gfgtdf> celmin: CAMPAIGN_LOW
20160815 19:30:43< celmin> gfgtdf: Oh, that doesn't count then.
20160815 19:32:05< gfgtdf> celmin: sure it does, at least for my point, which was about reladong of the game config in the new gui2 create dialog.
20160815 19:32:30< celmin> My point was that campaign_defines are, in general, not multiplayer-friendly.
20160815 19:33:05< celmin> Your point is a good one though. One more point against merging the two dialogs, in my opinion.
20160815 19:38:32< gfgtdf> celmin: the alterantive yould be to require wml to abhe teh configure_relevalnt wml outside of proprocessor defienes so that we cna load that wml after the configur dialog.
20160815 19:40:08< celmin> That'd be a lot of work.
20160815 19:40:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 19:41:06< gfgtdf> celmin: you mean on the wml or c++ side ?
20160815 19:41:27< celmin> Probably both, but more the C++ side.
20160815 19:48:55-!- edgrey [~edgrey@178.205.2.70] has quit [Quit: Konversation terminated!]
20160815 19:53:46-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20160815 20:04:08< gfgtdf> vultraz: is there a reason why teh gui1 scrren is shown after the gui2 create screen ?
20160815 20:11:41-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 20:28:58-!- Samual [~Samual@xonotic/core-team/Samual] has quit [Ping timeout: 250 seconds]
20160815 20:29:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 20:29:19-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 20:29:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 20:29:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 20:34:58-!- horrowind [~Icedove@2a02:810a:83c0:404:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160815 20:37:15-!- Samual [~Samual@xonotic/core-team/Samual] has joined #wesnoth-dev
20160815 20:45:18-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20160815 20:54:14-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160815 20:54:58-!- Samual [~Samual@xonotic/core-team/Samual] has quit [Ping timeout: 250 seconds]
20160815 20:55:33-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160815 20:55:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 20:56:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 20:57:06-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Client Quit]
20160815 20:58:12-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 48.0/20160726073904]]
20160815 21:02:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 21:02:41-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160815 21:07:51-!- Samual [~Samual@2601:547:1000:86f:cd8c:30ef:cc2e:6395] has joined #wesnoth-dev
20160815 21:07:51-!- Samual [~Samual@2601:547:1000:86f:cd8c:30ef:cc2e:6395] has quit [Changing host]
20160815 21:07:51-!- Samual [~Samual@xonotic/core-team/Samual] has joined #wesnoth-dev
20160815 21:12:44-!- Samual [~Samual@xonotic/core-team/Samual] has quit [Ping timeout: 250 seconds]
20160815 21:24:21-!- Samual [~Samual@xonotic/core-team/Samual] has joined #wesnoth-dev
20160815 21:31:42-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160815 21:34:32-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Client Quit]
20160815 21:44:03-!- Appleman1234 [~Appleman1@KD036012022057.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20160815 22:22:35-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20160815 22:24:40-!- Appleman1234 [~Appleman1@KD036012018092.au-net.ne.jp] has joined #wesnoth-dev
20160815 22:45:55-!- ancestral [~ancestral@209.181.254.220] has joined #wesnoth-dev
20160815 23:08:07< vultraz> gfgtdf, celmin: hm.  i didn't observe the config reloading when selecting a new entry
20160815 23:09:15< celmin> I would assume it only happens when switching between LoW and something else.
20160815 23:09:19< celmin> (Or other MP campaigns.)
20160815 23:09:22< vultraz> did not either
20160815 23:09:42< celmin> I haven't tried it myself though.
20160815 23:09:46< vultraz> well, I think it happened once
20160815 23:10:24< vultraz> anyway, any feedback?
20160815 23:10:39-!- ancestral [~ancestral@209.181.254.220] has quit [Ping timeout: 264 seconds]
20160815 23:11:28-!- ancestral [~ancestral@167.sub-174-219-134.myvzw.com] has joined #wesnoth-dev
20160815 23:12:46< celmin> Like I said, I haven't tried it yet,
20160815 23:13:35< vultraz> i thought you were looking at the code
20160815 23:13:47< celmin> What made you think that?
20160815 23:13:58< celmin> I think I looked at the code a few hours ago though.
20160815 23:14:09< celmin> I would've said something then if I'd noticed anything.
20160815 23:14:54< vultraz> ah
20160815 23:14:55< vultraz> ok
20160815 23:17:09-!- ancestral [~ancestral@167.sub-174-219-134.myvzw.com] has quit [Quit: i go nstuf kthxbai]
20160815 23:21:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160815 23:22:24-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Ping timeout: 258 seconds]
20160815 23:25:00-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has joined #wesnoth-dev
20160815 23:25:03< gfgtdf> 20160815 23:08:07< vultraz> gfgtdf, celmin: hm.  i didn't observe the config reloading when selecting a new entry
20160815 23:25:46< gfgtdf> vultraz: thats becasue in your version the prepare_... functiosn dotn dont really load the scneario
20160815 23:26:02< vultraz> hm?
20160815 23:26:06< gfgtdf> vultraz: i'll do  a commit wait
20160815 23:26:15< celmin> ...
20160815 23:34:12< vultraz> why "..."?
20160815 23:35:34-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20160815 23:36:08-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160815 23:40:27-!- irker191 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160815 23:40:27< irker191> wesnoth: gfgtdf wesnoth:master 60caf25d415a / src/ (3 files in 2 dirs): gui2 mp create improvements https://github.com/wesnoth/wesnoth/commit/60caf25d415ae29cd99862dad5703f7d4cbfa16f
20160815 23:40:54-!- gfgtdf [~chatzilla@x4e368e1d.dyn.telefonica.de] has joined #wesnoth-dev
20160815 23:40:59< gfgtdf> vultraz: ^
20160815 23:41:24< gfgtdf> vultraz: this also fixed the sewttignss defulat snot beeing updated
20160815 23:41:35< vultraz> ty :D
20160815 23:42:16< celmin> I'll try it out soon, then.
20160815 23:42:25< celmin> Maybe even fiddle with a low-res version.
20160815 23:42:46< gfgtdf> vultraz: but you'll now get the loadignscrreen issue that i noted earlier.
20160815 23:43:48< gfgtdf> vultraz: the point is that for some scnearios, speciall mp campaign but also wml-heavy mainle scneario liek a new land, the dies definition is guarded by a preprocessor macoro so the conif needs to be reloaded to display the settings whcih si annyoing
20160815 23:44:20< vultraz> hm
20160815 23:44:27< vultraz> im only seeing that happen with some scenarios
20160815 23:44:55< gfgtdf> vultraz: yes it onyl happens with scneairos that use prorocessor macors to guard their content
20160815 23:45:08< gfgtdf> vultraz: whcih is afaik a new lang and one other scenarios
20160815 23:45:18< gfgtdf> vultraz: and LoW campaign ofcourse
20160815 23:45:25< vultraz> any way to delegate that to a background thread?
20160815 23:46:32< gfgtdf> vultraz: well even if its done in another thread, that doenst change that the settings cannot be displayed immidiateley
20160815 23:46:47< vultraz> also, now if i select Random Maps, it says no sides found :/
20160815 23:47:06< gfgtdf> vultraz: yes previous it crashed in that case
20160815 23:47:10< vultraz> hm
20160815 23:47:20< gfgtdf> vultraz: thats becasue random maps are still not implemented preperly
20160815 23:47:35< gfgtdf> vultraz: you always call engine::prepare_for_scenario
20160815 23:47:42< vultraz> ah, right
20160815 23:47:43< vultraz> ok
20160815 23:47:59< gfgtdf> vultraz: the gui1 code called prepare_fpor_capaign, prpare_for_scenario, prepare_fpor_other depended on the type
20160815 23:48:04< vultraz> yeah
20160815 23:48:15< vultraz> but still, I dunno what to do here :/
20160815 23:48:24< vultraz> the loading issue is something we can't have
20160815 23:48:33< celmin> I don't think it's such a bad thing really.
20160815 23:48:49< celmin> It's not ideal, but it's not terrible either.
20160815 23:49:27< gfgtdf> celmin: well if you for exampel navigate to entry 5 with arrow keys, and entry 2 uses a custom define then youll get 2 loadingascrren (one form 1 to 2 and once form 2 to 3)
20160815 23:50:23< vultraz> also the gui1 version had no loading screens
20160815 23:50:25< vultraz> so it's a step backwards
20160815 23:50:48< gfgtdf> vultraz: the gui1 version had teh loadingscreen after the screte screen, (before th emp configure screen)
20160815 23:50:59< gfgtdf> s/screte/screte
20160815 23:51:05< vultraz> hm
20160815 23:51:05< celmin> create
20160815 23:51:19< gfgtdf> if you chose a scnerio that uses a cusotm define
20160815 23:52:05< vultraz> i could make the settings tab inaccessible until the end of create, but that would defeat the purpose of having it here
20160815 23:53:20< gfgtdf> vultraz: i reccomdne to change all those [multiplayer] to move all the values that are important for mp setting out if the #ifdef, this woudol work for all [multiplayer] but we still need to think about a way om how to handle it for campaigns
20160815 23:53:55< gfgtdf> vultraz: so that we onyl need the 'full' scneario after the mp create&configure screen
20160815 23:54:06< celmin> I don't think eliminating the use of the defines is a valid solution.
20160815 23:54:48< gfgtdf> celmin: we dont eleimiate that
20160815 23:54:59< gfgtdf> celmin: we just move ~4 lines in each of those wml files
20160815 23:55:08< celmin> Eh?
20160815 23:55:47< vultraz> celmin: he's saying we move the data necessary to display out of the ifdefs
20160815 23:56:02< celmin> But that won't actually help.
20160815 23:56:10< celmin> We can't force addons to upgrade themselves.
20160815 23:56:32< celmin> Fixing this problem by altering the WML is useless. It's not a fix at all.
20160815 23:56:50< vultraz> looks like SP just needs a list of scenarios
20160815 23:56:55< vultraz> er, campaigns
20160815 23:57:20< gfgtdf> celmin: defeines in [modification] is a 1.13 feature so we dont need to worry about that, also it rreally wount be teh firt time we make changes to wml liek that
20160815 23:57:55< celmin> It's not as if no-one makes addons for 1.13, so saying you don't need to worry about it is a bit...
20160815 23:58:12< celmin> Anyway, there's still LoW to worry about, right?
20160815 23:58:31< celmin> And any other add-on MP campaigns.
--- Log closed Tue Aug 16 00:00:10 2016