--- Log opened Wed Aug 26 00:00:00 2015
20150826 00:01:10-!- framling [~user@c-50-186-43-50.hsd1.or.comcast.net] has joined #wesnoth-dev
20150826 00:18:54-!- TC01 [~quassel@london.acm.jhu.edu] has quit [Ping timeout: 260 seconds]
20150826 00:21:46-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150826 00:22:32-!- TC01 [~quassel@london.acm.jhu.edu] has joined #wesnoth-dev
20150826 00:34:02-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150826 00:37:12-!- BeoXTC_ [3eb2b12a@gateway/web/cgi-irc/kiwiirc.com/ip.62.178.177.42] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
20150826 00:45:36-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150826 00:55:53-!- Appleman1234 [~Appleman1@KD119104014020.au-net.ne.jp] has quit [Ping timeout: 256 seconds]
20150826 01:24:25-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150826 01:24:25-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150826 01:24:25-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150826 01:37:14-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20150826 01:53:31-!- Appleman1234 [~Appleman1@KD119104015216.au-net.ne.jp] has joined #wesnoth-dev
20150826 02:03:19< celticminstrel> mattsc: I did that, didn't I?
20150826 02:03:56< mattsc> you did?  Hmm, why didn’t I get an email about it then?
20150826 02:04:12< celticminstrel> I dunno, maybe it doesn't email you when I push new commits to a PR.
20150826 02:04:22< celticminstrel> I probably should've commented as well to alert you.
20150826 02:04:27-!- irker650 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150826 02:05:25< mattsc> Hmm, yeah, it is changed.  Strange, usually I get emails about any changes or comments to PRs (so there’s no need to alert me)
20150826 02:05:47< mattsc> celticminstrel: sorry for that, I should have checked earlier.
20150826 02:05:55< celticminstrel> That's okay.
20150826 02:06:06< mattsc> I probably won’t have time for this tonight, but will check it out and likely merge it tomorrow.
20150826 02:06:15< celticminstrel> Alright.
20150826 02:06:19< mattsc> Too much work and family stuff going on tonight.  :P
20150826 02:11:15-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Ping timeout: 245 seconds]
20150826 02:12:56-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20150826 02:13:05-!- sailorsw_ [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150826 02:16:43-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has quit [Ping timeout: 244 seconds]
20150826 02:19:56-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150826 02:23:12-!- sailorsw_ [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has quit [Ping timeout: 255 seconds]
20150826 02:35:03< shadowm> wedge009: It didn't seem to make a difference to me. Maybe I'll try again later.
20150826 02:41:56< mattsc> celticminstrel: ugh, I just tried to do this quickly, but when I change my local repository to PR464, I get the version from 6 days ago, not the one I can see on the github website.
20150826 02:42:28< celticminstrel> Uh. Did you do git fetch/pull/whatever?
20150826 02:43:05< mattsc> probably not …  my brain’s too tired to work at the moment
20150826 02:43:57< celticminstrel> I see. Well it's hardly urgent.
20150826 02:44:20< mattsc> and the github app doesn’t give me a button that I dare to press  :P
20150826 02:44:44< mattsc> Btw, I did pull master, yes, but that’s not it.
20150826 02:44:50< celticminstrel> It seems to me that some of the animations aren't fully utilizing the potential of progressive strings.
20150826 02:44:59< celticminstrel> At least in 1.12. 
20150826 02:45:49< celticminstrel> For example, the shaman has "units/elves-wood/shaman.png:225,units/elves-wood/shaman-attack2.png:225,units/elves-wood/shaman.png:100" in its attack animation. It could be shortened to "units/elves-wood/[shaman,shaman-attack2,shaman].png:[225*2,100]", I believe.
20150826 02:46:47< celticminstrel> Possibly even "units/elves-wood/shaman[,-attack2,].png:[225*2,100]", but I'm not sure if empty elements are supported.
20150826 03:02:43-!- irker494 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150826 03:02:43< irker494> wesnoth: Celtic Minstrel wesnoth:master 113cd7061627 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: XCode: Copy translations to target if they have been built http://git.io/vsFVG
20150826 03:02:43< irker494> wesnoth: mattsc wesnoth:master f8dc5d027c30 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Merge pull request #464 from CelticMinstrel/trans http://git.io/vsFVZ
20150826 03:04:49< mattsc> celticminstrel: ^  I assume this pinged you already, but just in case
20150826 03:04:59< celticminstrel> It did.
20150826 03:05:19< celticminstrel> I'm set up to be pinged by several things, including celticminstrel, celmin, celtic, and minstrel.
20150826 03:05:28< mattsc> Thanks for doing this.  The only question I have and am undecided about is whether we really want the verbose flag
20150826 03:05:53< celticminstrel> I'm not sure either. The verbose flag does mean it'll write to your log whenever something is actually updated.
20150826 03:05:57-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 255 seconds]
20150826 03:05:58< mattsc> I have no real preference for either.
20150826 03:06:07< celticminstrel> As it does with the Copy Files stage, if I recall correctly.
20150826 03:06:27< mattsc> Right, something like that.
20150826 03:06:34< celticminstrel> Though the Copy Files stage doesn't recursively list directories being copied.
20150826 03:06:51< mattsc> Right.
20150826 03:06:54< mattsc> Anyways, I have to be off.  TTYL.
20150826 03:07:00< celticminstrel> Bye.
20150826 03:07:11-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20150826 03:07:38-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150826 03:10:00< celticminstrel> Also, the behaviour of leading zeros in numbers in progressive strings is weird and counterintuitive.
20150826 03:15:23< celticminstrel> I'm looking at the examples on the wiki ( http://wiki.wesnoth.org/AnimationWML#Range_expansion )
20150826 03:17:00< celticminstrel> "07~11" makes sense, "098~100" should logically expand to "098,099,100", "003~1" looks ill-formed but does basically make sense, "0098~100" also looks ill-formed but should really expand to "0098,0099,0100", and "01~100" makes sense.
20150826 03:18:59< iceiceice> celticminstrel, i never noticed this
20150826 03:19:54< celticminstrel> Which, the weird behaviour of leading zeros, or the lack of using progressive strings for full filenames rather than just numbers?
20150826 03:20:03< iceiceice> yeah that doesn't make sense to me either
20150826 03:20:22< celticminstrel> It's very well defined, but counter-intuitive.
20150826 03:20:46< iceiceice> why is 0098-100 not 0098, 0099, 0100
20150826 03:21:02< celticminstrel> It's based entirely on the number of zeros.
20150826 03:21:21< iceiceice> i see
20150826 03:21:21< celticminstrel> ie, if it was "0007~100", then you'd end up with a list ending in "0098, 0099, 0100".
20150826 03:21:25< iceiceice> so its like a format specifier sort of
20150826 03:21:53< iceiceice> huh
20150826 03:22:31< celticminstrel> If it's a format specifier you want, I'd suggest something less confusing, eg "1~7:2" yields "01,02,03,04,05,06,07".
20150826 03:22:50< celticminstrel> Maybe I'll have a go at implementing that later.
20150826 03:23:18< celticminstrel> Right now I'm updating all the animations in my campaign to use start_time and duration instead of begin and end (and also progressive strings).
20150826 03:30:18< shadowm> "“098~100” would become “98,99,100”, " This example seems wrong? Is this really what Wesnoth does?
20150826 03:30:32< shadowm> The description right above the examples also contracts the example.
20150826 03:30:37< shadowm> contradicts
20150826 03:30:56-!- pydsigner is now known as pydsigner-vacati
20150826 03:31:11-!- pydsigner-vacati is now known as pydsigner-awaay
20150826 03:32:02-!- pydsigner-awaay is now known as pydsigner
20150826 03:32:24< celticminstrel> shadowm: I haven't actually tested it, so I'm not sure.
20150826 03:33:38-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150826 03:33:56-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 03:40:35-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150826 03:45:45-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20150826 03:46:27-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has joined #wesnoth-dev
20150826 03:53:17-!- sfan786 [~sfan786@c-98-239-230-28.hsd1.pa.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150826 03:55:15< celticminstrel> However, I don't think the examples and description contradict each other.
20150826 03:56:02< shadowm> "If either x or y (or both) are padded with leading zeros, then each number in the resultant list will be padded with leading zeros such that it is expressed in at least z+1 digits, where z is the quantity of leading zeros with which one or both endpoints are padded."
20150826 03:56:05< celticminstrel> In the first two examples and the last example, z=1, so the numbers are padded to at least two digits. In the other two examples, z=2, so the numbers are padded to at least three digits.
20150826 03:56:56< shadowm> Maths.
20150826 03:57:02 * shadowm shouldn't even try.
20150826 03:58:07< celticminstrel> Assuming it really is accurate (which I assume it is), it would break compatibility to change it now, though if very few add-ons relied on leading zeros it might not be a major issue.
20150826 03:59:00< celticminstrel> Even if it was changed like that though, I'm not sure it wouldn't be confusing. Less confusing, certainly, but perhaps still a little confusing.
20150826 04:17:25< shadowm> gfgtdf: wesnoth.random is not documented.
20150826 04:17:38< shadowm> At least I can't find it here: http://wiki.wesnoth.org/LuaWML
20150826 04:19:46< celticminstrel> I want a way to patch in resistances to existing movetypes. Any suggestions?
20150826 04:20:04< celticminstrel> (For people who invent a new damage type.)
20150826 04:21:14-!- oldlaptop [~quassel@50-107-94-87.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 244 seconds]
20150826 04:29:41-!- oldlaptop [~quassel@50-107-94-87.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150826 04:41:25-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150826 04:44:59-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20150826 04:54:05-!- iwaim [~iwaim@rasteenie.alib.jp] has quit [Quit: Tiarra 0.1: SIGTERM received; exit]
20150826 04:54:55-!- iwaim_ [~iwaim@rasteenie.alib.jp] has joined #wesnoth-dev
20150826 05:24:40-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Remote host closed the connection]
20150826 05:25:27-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 05:37:22-!- subliun [~quassel@139.216.140.254] has left #wesnoth-dev ["http://quassel-irc.org - Chat comfortably. Anywhere."]
20150826 05:38:59-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150826 05:39:34-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 05:41:59-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Client Quit]
20150826 05:42:26-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 05:43:58< celticminstrel> Has anyone other than vultraz looked at PR475 yet?
20150826 05:56:14-!- Shackra [~Jorge@186.177.2.148] has quit [Remote host closed the connection]
20150826 06:36:56< shadowm> I'll do it tomorrow or so, but I'm already suspicious of the approach here.
20150826 06:37:21< shadowm> Mostly because it's very explicitly OS X-specific.
20150826 06:37:40< celticminstrel> The problem itself is semi-OSX-specific.
20150826 06:38:29< shadowm> We agreed at the very beginning of this endeavor that the close button on all platforms was an issue to consider too, as well as Alt+F4 on Windows and Linux (at least with some WMs on the latter).
20150826 06:39:41< celticminstrel> If wedge009 is correct, Alt+F4 might be as simple as setting a Windows-specific default hotkey.
20150826 06:40:22< celticminstrel> As for the close button, I agree in principle, but the implementation was proving harder than I'd anticipated.
20150826 06:40:36< shadowm> I haven't seen wedge009's theory yet I think.
20150826 06:41:12< shadowm> However, I know this: Alt+F3 is not passed to Wesnoth by my WM, instead Wesnoth gets only Alt.
20150826 06:41:32< shadowm> Alt+F3 serves the same function as Alt+Space (IIRC) does on Windows, which is to bring up the window's menu.
20150826 06:41:45< celticminstrel> Yeah, that's Alt+Space.
20150826 06:41:48< shadowm> With this in mind, I doubt Alt+F4 would be passed to Wesnoth either.
20150826 06:42:16< celticminstrel> I'm not really sure.
20150826 06:43:28< celticminstrel> In the case of Cmd+Q though, the WM does pass it to Wesnoth, but Wesnoth doesn't pass it on to the event handlers asa keypress.
20150826 06:43:40< celticminstrel> ^as a
20150826 06:43:57< celticminstrel> Instead it triggers the menuitem, which then pushes an SDL_QUIT event onto the queue.
20150826 06:44:30< shadowm> Looks like OS X works completely differently from X11 in this regard.
20150826 06:44:56< celticminstrel> There may be another approach I could try... but it involves some sort of global variable.
20150826 06:45:38< shadowm> Just took a look to see whether xev (an X11 utility that logs raw X11 events) could see Alt+F4 and it doesn't either, it receives the WM_DELETE_WINDOW message first.
20150826 06:45:51< celticminstrel> I see.
20150826 06:48:01< shadowm> From a cursory glance at this SDLMain.m it's pretty obvious OS X allows finer control over this stuff for... some reason. On Windows and X11 we just have the WM's messages (which SDL translates to events for us) to go by.
20150826 06:48:47< shadowm> And it seems reasonable that the WM would not relay key press events involved in a WM control sequence.
20150826 06:49:07< celticminstrel> Well, Cmd+Q isn't a WM control sequence.
20150826 06:49:22< celticminstrel> It's the equivalent of Ctrl-Q.
20150826 06:49:22< shadowm> Alt+F4 is on Windows, for sure, and apparently on KWin as well.
20150826 06:49:33< celticminstrel> I don't believe the Mac has an equivalent of Alt+F4.
20150826 06:49:52< shadowm> Maybe the key presses will get relayed if the application chooses to ignore the event but I'd have to see.
20150826 06:50:05< shadowm> WEll, thus far I was led to believe Cmd+Q is equivalent to Alt+F4. :p
20150826 06:50:22< celticminstrel> That's probably just because of the way SDLMain handled it. :P
20150826 06:50:53< celticminstrel> I don't think it's strictly unreasonable for Cmd+Q to generate a quit event, mind you.
20150826 06:52:40-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150826 06:52:45< celticminstrel> It may not be a WM control sequence like Alt+F4 is, but Apple guidelines specify it should only be used for quit.
20150826 06:53:20< celticminstrel> It feels like sending it as a keypress simplifies the handling though. I dunno.
20150826 06:53:42< celticminstrel> (Though it also makes it rebindable, which is weird.)
20150826 06:53:49< shadowm> Exactly.
20150826 06:55:12< shadowm> A +22 lines / -22 lines overall diff doesn't seem to be simplifying things to me either.
20150826 06:55:32< celticminstrel> Well, compared to all the work I was doing to try to get it to work the other way. >_>
20150826 06:55:36< shadowm> Plus apparently this removes the handling of Escape on all platforms (?) for no obvious reason.
20150826 06:55:36< celticminstrel> But I have a few other ideas.
20150826 06:55:50< celticminstrel> It doesn't remove it, just shunts it into the hotkey system.
20150826 06:55:59< celticminstrel> (For Escape)
20150826 06:56:20< shadowm> Ah, right.
20150826 06:56:35< celticminstrel> So basically, Escape still works, and it even works in the editor, but it's also rebindable. I don't think that's as weird as Cmd+Q being rebindable, personally.
20150826 06:56:56-!- [Relic] [~Relic]@2602:306:33a3:6d30:65c2:c1ad:67b3:bc91] has quit [Quit: I press the magic X and all the weirdos go away!]
20150826 06:56:58< shadowm> By all means, that ought to be its own separate commit. :p
20150826 06:57:02< celticminstrel> Probably.
20150826 06:57:20< celticminstrel> If I could get SDL_QUIT somehow shunted into the hotkey handling system...
20150826 06:57:37< celticminstrel> As a special case hard-coded to HOTKEY_QUIT_TO_DESKTOP...
20150826 07:01:43< shadowm> I can't say I'd trust users with this power.
20150826 07:01:50< celticminstrel> Eh?
20150826 07:02:11< shadowm> This is not the kind of stuff people should be able to configure, in my view.
20150826 07:04:04< shadowm> I can't say I approve of removing too much power from users like Apple or Gnome do, but it's necessary to draw a line somewhere to keep programmers and tech support people sane.
20150826 07:04:35< celticminstrel> Does that apply to Escape or just to Cmd+Q?
20150826 07:04:48< shadowm> Both, really.
20150826 07:05:14< celticminstrel> Eh.
20150826 07:05:23< shadowm> "I was messing with the game preferences the other day and now the game quits every time I press the mouse button help" (user remapped Mouse0Button1 to Quit)
20150826 07:05:42< celticminstrel> I suspect that's already possible.
20150826 07:06:20< shadowm> Just to give a very extreme example. Odds are mouse handling like that only works on the map viewport.
20150826 07:06:42< shadowm> Sure, but does that mean we should keep giving people the ability to shoot themselves on the foot and blame us later?
20150826 07:07:22< celticminstrel> Perhaps not.
20150826 07:08:04< shadowm> If it makes the code simpler for mysterious reasons, I'd rather add the ability to set some hotkeys to read-only so they can't be accidentally changed and aren't read to or written to WML.
20150826 07:08:15< celticminstrel> I'm not against that.
20150826 07:08:21< shadowm> *read from
20150826 07:09:37< celticminstrel> By changing Escape to a hotkey, it now "magically" works in the editor, something that I seem to recall vultraz was attempting to do at one point. (Possible minor downside is it also works at the titlescreen, but I fixed that by making the titlescreen handle Quit to Desktop instead of Quit to Titlescreen.)
20150826 07:11:34< shadowm> Now, considering treating a non-key event like SDL_QUIT as a hotkey suggests a severe defficiency in Wesnoth's event handling design.
20150826 07:11:58< celticminstrel> Are you saying I shouldn't do that?
20150826 07:12:07< shadowm> Unlike mouse or joystick events, it does not necessarily have to originate from user input.
20150826 07:12:31< celticminstrel> I suppose that's true. On OSX it could originate from an AppleScript.
20150826 07:12:35< shadowm> For example, on Linux SDL generates this event on SIGINT or SIGTERM to Wesnoth's process.
20150826 07:13:23< celticminstrel> But then how else are we supposed to handle close button or Alt+F4?
20150826 07:13:47< celticminstrel> I don't think it's unreasonable to show confirmation on SIGINT. SIGTERM on the other hand, that seems a bit silly.
20150826 07:13:59< celticminstrel> A lot silly.
20150826 07:14:07< shadowm> Make it so this is event is handled as a thing of its own rather than shoehorn it into the hotkeys code?
20150826 07:15:03< shadowm> I'd rather not see any confirmation on SIGINT, myself. This signal is sent when pressing Ctrl+C on the process' parent console, which I do a lot when testing for obvious reasons.
20150826 07:15:45< celticminstrel> Fair enough, but from the sound of it, it may be quite difficult to show confirmation on Alt+F4 but not on SIGINT.
20150826 07:15:47< shadowm> All I'm saying here in this regard is that Wesnoth is full of ugly hacks and it's crumbling apart due to their sheer weight accumulating over time.
20150826 07:16:00< celticminstrel> I suppose if you installed a signal handler of your own to override SDL's.
20150826 07:16:26< celticminstrel> I thought SIGTERM was uncatchable. Maybe I was confusing it with SIGKILL.
20150826 07:16:30< shadowm> We can keep adding them to 1.12 if needed, but moving forward from there we need cleaner code if we don't want the game to catch fire on its own and die due to there being no developers left able to understand the code.
20150826 07:16:57< shadowm> All signals other than SIGKILL and SIGSTOP can be handled, IIRC this is defined by POSIX.
20150826 07:19:17< shadowm> "The SIGKILL and SIGSTOP signals shall not be added to the signal mask using this mechanism; this restriction shall be enforced by the system without causing an error to be indicated." Yep.
20150826 07:19:25< celticminstrel> Okay, so it's relatively easy to change the event handler to invoke a callback instead of throwing an exception when receiving SDL_QUIT (I did that in the other branch).
20150826 07:20:42< celticminstrel> And I can make controller_base override that callback to call cmd_exec->execute_command(HOTKEY_QUIT_TO_DESKTOP).
20150826 07:21:19< celticminstrel> That's basically what I meant by shunting it into the hotkey system.
20150826 07:22:44< celticminstrel> I've waffled on what to do when a dialog is onscreen, but I've come to the conclusion both that it's simplest to just do nothing and that it kind of does make sense to ignore a quit request if a modal dialog is up.
20150826 07:24:25< shadowm> Okay, I thought you meant forcing it into the hotkey table and stuff.
20150826 07:28:27< shadowm>     setenv ("SDL_VIDEO_ALLOW_SCREENSAVER", "1", 1);
20150826 07:28:36< shadowm> Why do we do this on OS X and only OS X... :p
20150826 07:28:48< celticminstrel> I have absolutely no clue.
20150826 07:29:37< shadowm> Because often I've left Wesnoth running and then come back to find my monitor still on, which is annoying.
20150826 07:30:30< celticminstrel> The SDLMain file is in fact included with SDL, though this one has clearly been customized a little.
20150826 07:31:14< shadowm> I imagine at some point some OS X packager found it necessary for some reason, like I was planning to override SDLMain on Windows for 1.13.2.
20150826 07:32:14< celticminstrel> Okay, I've implemented this way, but I think I'm going to put off testing until tomorrow. For now, Escape is still rebindable.
20150826 07:33:13< celticminstrel> Perviously it was special-cased in command_executor, perhaps special-casing it in controller_base would be more appropriate.
20150826 07:33:17< celticminstrel> ^Previously
20150826 07:34:08< irker494> wesnoth: Wedge009 wesnoth:master f05557a7d9e9 / data/ (3 files in 2 dirs): Spelling corrections http://git.io/vsbGj
20150826 07:34:10< irker494> wesnoth: Wedge009 wesnoth:master e37ee6a1923a / po/ (wesnoth-help/wesnoth-help.pot wesnoth-lib/wesnoth-lib.pot): Spelling corrections http://git.io/vsbZe
20150826 07:34:12< irker494> wesnoth: Charles Dang wesnoth:master 66b6ca39a1e7 / / (5 files in 4 dirs): Merge pull request #476 from Wedge009/Spelling_Corrections http://git.io/vsbZv
20150826 07:34:16< celticminstrel> I might try that. I'll also look into installing signal handlers for SIGINT/SIGTERM.
20150826 07:34:23< celticminstrel> Since you feel they should insta-quit.
20150826 07:34:39< celticminstrel> Good night.
20150826 07:34:47-!- celticminstrel is now known as celmin|slee[
20150826 07:34:51-!- celmin|slee[ is now known as celmin|sleep
20150826 07:38:30-!- iwaim_ [~iwaim@rasteenie.alib.jp] has quit [Remote host closed the connection]
20150826 07:39:04-!- iwaim [~iwaim@rasteenie.alib.jp] has joined #wesnoth-dev
20150826 07:41:11-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 252 seconds]
20150826 07:48:11< Soliton> the default action for SIGINT/SIGTERM is already (instant) process termination.
20150826 07:48:44< celmin|sleep> It sounds like SDL may override that? I dunno.
20150826 07:49:11< celmin|sleep> I'll find out when I start testing.
20150826 07:49:46< Soliton> what sounds like that? AFAIK ^C terminates wesnoth just fine.
20150826 08:10:55< shadowm> The point is that this particular aspect would change with the proposed changes to how the SDL_QUIT event is handled, since that's how Wesnoth finds out that SIGINT or SIGTERM have been issued (SDL's SIGINT/SIGTERM handler gnerates the event but apparently doesn't provide a way to tell *why* it was generated).
20150826 08:17:48< Soliton> so SDL2 does install signal handlers?
20150826 08:17:48-!- joet [~joet@host86-163-217-248.range86-163.btcentralplus.com] has joined #wesnoth-dev
20150826 08:18:50< Soliton> maybe it can be convinced not to then or remove the signal handlers afterwards.
20150826 08:18:52< shadowm> SDL 1.2 and SDL 2 both do the same thing, yes.
20150826 08:21:48< shadowm> It's been ages since I last messed with POSIX signal handling but I believe we should be able to override SDL's handlers. Looking at SDL 2's source there also seems to be a way to tell it to not install them in the first place, unlike 1.2.
20150826 08:22:31< shadowm> Since version 2.0.4, at least.
20150826 08:23:54< shadowm> https://wiki.libsdl.org/SDL_HINT_NO_SIGNAL_HANDLERS -- not quite sure exactly how this is supposed to be used with https://wiki.libsdl.org/SDL_SetHint , but I assume it's prior to calling SDL_Init().
20150826 08:24:22< shadowm> And 2.0.4 isn't even out yet, it turns out.
20150826 08:28:28-!- tomreyn [~tomreyn@p4FF5ACB9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150826 08:43:22-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 260 seconds]
20150826 08:47:57-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150826 09:02:13-!- mjs-de [~mjs-de@x4db580a5.dyn.telefonica.de] has joined #wesnoth-dev
20150826 09:18:08-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150826 09:23:16-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has quit [Quit: i go nstuf kthxbai]
20150826 09:24:15-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 09:25:49-!- c74d3 [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20150826 09:28:15-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Ping timeout: 241 seconds]
20150826 09:28:16-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20150826 09:28:16-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has quit [Excess Flood]
20150826 09:28:17-!- wedge010 is now known as wedge009
20150826 09:29:24-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has joined #wesnoth-dev
20150826 09:39:10-!- sailorswift [~sailorswi@c-50-156-105-171.hsd1.ca.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150826 10:06:55-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150826 10:35:00-!- irker494 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150826 10:36:25-!- new_one [~new_one@2604:a880:1:20::22e:d001] has quit [K-Lined]
20150826 10:39:38-!- tomreyn [~tomreyn@p4FF5ACB9.dip0.t-ipconnect.de] has quit [Quit: tomreyn]
20150826 10:40:08-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150826 10:47:09-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20150826 10:47:23-!- new_one [~new_one@2604:a880:1:20::22e:d001] has joined #wesnoth-dev
20150826 10:55:03-!- new_one [~new_one@2604:a880:1:20::22e:d001] has quit [K-Lined]
20150826 11:13:03-!- prkc [~prkc@51B77E5A.dsl.pool.telekom.hu] has joined #wesnoth-dev
20150826 11:16:03-!- new_one [~new_one@2604:a880:1:20::22e:d001] has joined #wesnoth-dev
20150826 11:50:33-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150826 11:51:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150826 11:51:09-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 11:51:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20150826 11:56:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150826 11:59:11< Ravana_> forums seem to be down
20150826 12:00:50-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150826 12:04:09-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150826 12:04:30-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 13:03:38-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150826 13:05:41-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 252 seconds]
20150826 13:05:42-!- wedge010 is now known as wedge009
20150826 13:09:19-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150826 13:18:24-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Quit: Quitte]
20150826 13:21:43-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150826 13:55:06-!- tao [9a750101@gateway/web/freenode/ip.154.117.1.1] has joined #wesnoth-dev
20150826 13:55:30-!- tao is now known as Guest70491
20150826 13:57:39-!- Guest70491 [9a750101@gateway/web/freenode/ip.154.117.1.1] has quit [Client Quit]
20150826 14:30:33-!- sfan786 [~sfan786@c-24-131-93-63.hsd1.pa.comcast.net] has joined #wesnoth-dev
20150826 14:39:06-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20150826 14:52:04-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150826 14:52:05-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150826 14:52:05-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150826 15:00:57-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has quit [Quit: i go nstuf kthxbai]
20150826 15:27:33< celmin|sleep> It looks like SIGTERM generates SDL_QUIT but SIGINT doesn't...
20150826 15:28:01< celmin|sleep> At least on OSX.
20150826 15:29:25< celmin|sleep> Though maybe it's different if launched from Terminal. I dunno.
20150826 15:37:22-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150826 15:46:48-!- celmin|sleep is now known as celticminstrel
20150826 15:49:18-!- [Relic] [~Relic]@2602:306:33a3:6d30:40f7:44ab:8f7a:3a45] has joined #wesnoth-dev
20150826 16:03:44-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 246 seconds]
20150826 16:04:05-!- sfan786 [~sfan786@c-24-131-93-63.hsd1.pa.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150826 16:16:59-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150826 16:21:21< celticminstrel> PR updated.
20150826 16:29:09-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150826 16:33:27-!- sfan786 [~sfan786@c-24-131-93-155.hsd1.pa.comcast.net] has joined #wesnoth-dev
20150826 16:46:02-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150826 16:53:13-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has quit [Ping timeout: 250 seconds]
20150826 17:09:45-!- sfan786 [~sfan786@c-24-131-93-155.hsd1.pa.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150826 17:30:25-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150826 18:00:50< celticminstrel> Anyone else getting a lot of preprocessor warnings when starting a local "MP" game? http://pastebin.com/kktZJaVB
20150826 18:08:55-!- prkc [~prkc@51B77E5A.dsl.pool.telekom.hu] has quit [Ping timeout: 244 seconds]
20150826 18:17:43-!- joet [~joet@host86-163-217-248.range86-163.btcentralplus.com] has quit [Ping timeout: 244 seconds]
20150826 18:23:41< celticminstrel> Does no-one else think it's weird that Ctrl+Q means "quit to titlescreen/lobby" rather than "quit to desktop"?
20150826 18:28:49-!- gfgtdf [~chatzilla@f054128111.adsl.alicedsl.de] has joined #wesnoth-dev
20150826 18:31:12-!- gfgtdf [~chatzilla@f054128111.adsl.alicedsl.de] has quit [Client Quit]
20150826 18:32:46-!- prkc [~prkc@51B77E5A.dsl.pool.telekom.hu] has joined #wesnoth-dev
20150826 18:48:25-!- prkc [~prkc@51B77E5A.dsl.pool.telekom.hu] has quit [Remote host closed the connection]
20150826 18:51:16-!- gfgtdf [~chatzilla@f054128111.adsl.alicedsl.de] has joined #wesnoth-dev
20150826 18:51:54< gfgtdf> celticminstrel: well i have windows here, but if Alt-F4 would default to 'back to titlescreen' then that would be weird.
20150826 19:01:21< celticminstrel> Since you have Windows, Ctrl+Q does indeed mean "quit to titlescreen", right? Does that seem weird, or not?
20150826 19:03:21< gfgtdf> celticminstrel: on windows Ctrl-Q is just a hotkey as any other, so no its doesn't seem weird.
20150826 19:03:35< celticminstrel> For me, it sort of works with Cmd+W for "quit to titlescreen", though in the editor I really want Cmd+W to be "close map"...
20150826 19:03:58< celticminstrel> Ctrl+Q has no special meaning on OSX either.
20150826 19:04:19< celticminstrel> But since it's the roughly equivalent to Cmd+Q, it feels weird to me that it wouldn't quit to desktop on Windows.
20150826 19:04:27< celticminstrel> ^-the
20150826 19:05:23< gfgtdf> celticminstrel: does appl also have a ctrl key ?
20150826 19:05:35< celticminstrel> Yes.
20150826 19:05:49< gfgtdf> celticminstrel: so it has 2 diferent keys cmd and ctrl ?
20150826 19:05:52< celticminstrel> Yes.
20150826 19:05:55< celticminstrel> And Alt.
20150826 19:06:27< celticminstrel> However Cmd on OSX is used basically how Ctrl is on other OSes.
20150826 19:06:37< celticminstrel> ie, the primary accelerator key
20150826 19:07:11< celticminstrel> Ctrl on the other hand isn't used as much. Ctrl-click means right-click, as well.
20150826 19:10:16< celticminstrel> ...speaking of which, Wesnoth doesn't honour that equivalence.
20150826 19:11:57< gfgtdf> on windows i have Alt which is mainly used for navigaing though Menus. and the Ctrl key which is mainly used for random hotkeys like copypase. (And AltGr which has the same effect as pressing strg and alt at the same time)
20150826 19:12:25< gfgtdf> i actuyl dont know for which thing u use Ctrol other than ctrl-C clrtV for copypaste
20150826 19:12:32< gfgtdf> oher things
20150826 19:12:48< celticminstrel> Those shortcuts are shown in the program menus. :P
20150826 19:12:50< celticminstrel> OSX uses Alt roughly like AltGr, for typing special characters.
20150826 19:14:25< gfgtdf> celticminstrel: then which ise used on OSX for navigating though menus ?
20150826 19:14:41< celticminstrel> There's no such thing.
20150826 19:14:43< celticminstrel> ...well.
20150826 19:14:52< celticminstrel> There is ...what was it again...
20150826 19:16:12< celticminstrel> Ah, Ctrl+F2 focuses the menubar, similar to pressing Alt on Windows.
20150826 19:16:23< celticminstrel> Though it might be disabled by default.
20150826 19:16:31< celticminstrel> You can then navigate with the arrow keys.
20150826 19:16:48< gfgtdf> celticminstrel: y that makes sense
20150826 19:17:16< celticminstrel> No single-key mnemonics like Windows, but starting to type a menu(item) name selects the first match.
20150826 19:20:44-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Ping timeout: 244 seconds]
20150826 19:21:17-!- irker607 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150826 19:21:17< irker607> wesnoth: gfgtdf wesnoth:master 982113353317 / src/savegame.cpp: Fix savegame compability for [advance]->[advancement] patch http://git.io/vspVJ
20150826 19:21:19-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20150826 19:22:17< gfgtdf> ?? when i tried to push ot gave me an error but it seems like the commit was sended :s
20150826 19:23:44-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20150826 19:28:23< gfgtdf> zookeeper: what was teh reason agains the konrad loyalty easer egg ?
20150826 19:30:14-!- Kwandulin [~Miranda@p5B0094F6.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150826 19:31:07< gfgtdf> zookeeper: you have an opinion on https://gna.org/bugs/?22974 ?
20150826 19:33:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150826 19:34:21-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20150826 19:37:18-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150826 19:39:31-!- klemvor [~klemvor@adsl-ull-169-103.40-151.net24.it] has joined #wesnoth-dev
20150826 19:41:46-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Client Quit]
20150826 19:43:19-!- esr [~esr@wesnoth/developer/esr] has quit [Quit: WeeChat 1.0.1]
20150826 19:44:54-!- esr [~esr@wesnoth/developer/esr] has joined #wesnoth-dev
20150826 19:45:24-!- BeoXTC_ [3eb2b12a@gateway/web/cgi-irc/kiwiirc.com/ip.62.178.177.42] has joined #wesnoth-dev
20150826 19:49:14< klemvor> hi everybody
20150826 19:50:22< klemvor> I would like to start contributing in the maintenance of the Python utilities of wesnoth
20150826 19:50:36< zookeeper> gfgtdf, i don't have anything against the modifications thing. i don't really understand the "resource" concept in your PR though.
20150826 19:50:57< zookeeper> is it just that any tags written there will be copied verbatim into every scenario?
20150826 19:50:59< gfgtdf> zookeeper: ah sry wrong bug number
20150826 19:51:05< zookeeper> oops
20150826 19:51:33< gfgtdf> zookeeper: i mean this one http://gna.org/bugs/?23769 which is easier to implement.
20150826 19:52:02< zookeeper> gfgtdf, yes, allowing [event]s inside [campaign] is something i've kinda wanted for a long time
20150826 19:52:30< gfgtdf> zookeeper: ok i'll do that then (if its as easy as ai think).
20150826 19:52:48< gfgtdf> zookeeper: abd the other question aout the Httt easter egg ?
20150826 19:53:58-!- TC01_ [~quassel@2601:14d:4000:4f79:6e62:6dff:fe9f:6b86] has joined #wesnoth-dev
20150826 19:55:11-!- TC01_ [~quassel@2601:14d:4000:4f79:6e62:6dff:fe9f:6b86] has quit [Remote host closed the connection]
20150826 19:56:20< gfgtdf> zookeeper: about*
20150826 20:01:59< zookeeper> gfgtdf, there's several reasons really. it invites gaming it, but then it's also unreliable (the 50% chance), and in practise it can seem pretty arbitrary
20150826 20:03:41< gfgtdf> zookeeper: hm the 50% chance can be removed.
20150826 20:03:51< gfgtdf> zookeeper: so you would remove that TODO from httt ?
20150826 20:04:00-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has joined #wesnoth-dev
20150826 20:04:09< gfgtdf> Elvish_Hunter: ^
20150826 20:04:18< gfgtdf> no sry
20150826 20:04:20< gfgtdf> klemvor: ^
20150826 20:04:34< zookeeper> yeah, i should remove it...
20150826 20:11:09< irker607> wesnoth: gfgtdf wesnoth:master 30dc78636bbf / src/saved_game.cpp: allow [event] (and [lua]) inside [campaign] http://git.io/vspdO
20150826 20:11:16< gfgtdf> zookeeper: ^
20150826 20:11:48-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Ping timeout: 264 seconds]
20150826 20:12:13< Elvish_Hunter> gfgtdf: so this allows defining global events for a campaign?
20150826 20:12:36-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20150826 20:13:03< gfgtdf> Elvish_Hunter: yes you can put [event] inside a campaign which will then be copied into even scenatio, similar to [event]a in [era]
20150826 20:14:06< gfgtdf> Elvish_Hunter: note that events with first_time_only=yes will be copied into evey new scenario even if tehy already fired in an earlier scenario so you need filters if you really want first_time_:only events here
20150826 20:14:24< gfgtdf> Elvish_Hunter: again: just liek [event]s in [era]
20150826 20:15:21< Elvish_Hunter> Very interesting. I guess that it'll be really useful for the common die events that every campaign has :-)
20150826 20:20:56< klemvor> sorry gfgtdf I'm not very accustomed to the irc slang, for what ^ stands for?
20150826 20:23:46< zookeeper> oh well, git borked my clone again...
20150826 20:24:57< gfgtdf> klemvor: it means 'look at the text above' here the i meant the 'Elvish_Hunter has joined ' notification
20150826 20:27:51< Ravana_> I am not sure all people have the join notifications displayed
20150826 20:28:30< klemvor> gfgtdf, Ravana: yes I have, I will talk to Elvish_Hunter ASAP
20150826 20:28:49-!- prkc [~prkc@51B77E5A.dsl.pool.telekom.hu] has joined #wesnoth-dev
20150826 20:34:55< zookeeper> so, i've done a mixed reset to get rid of some local commits, but that ended up with untold amounts of stuff as local changes. how do i undo that so that only the local changes i had prior remain?
20150826 20:35:22-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150826 20:35:33< zookeeper> i'm guessing "you can't"
20150826 20:35:53< gfgtdf> zookeeper: you can doa  giet reset hard but im not sure if that what you want
20150826 20:36:05< zookeeper> so i guess i have to somehow remember which one of these >3000 files had local changes
20150826 20:36:07< shadowm> You can't unless you go back to the tip of the commit set you discarded.
20150826 20:38:10< zookeeper> wait... i think it didn't actually modify any files (except the ones which the discarded commits affected)
20150826 20:41:35-!- prkc [~prkc@51B77E5A.dsl.pool.telekom.hu] has quit [Remote host closed the connection]
20150826 20:41:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150826 20:42:50< zookeeper> shadowm, and i take it that if i re-apply the patch which created that commit set, it won't do?
20150826 20:45:17< shadowm> That'd be equivalent to what I described. Also, the safer route is to stash your local changes prior to resurrecting the commits, then try to apply the stash.
20150826 20:48:54< zookeeper> trying to stash untracked files results in some warning about stuff possibly being deleted which i don't understand, so i can't do that
20150826 20:49:26< shadowm> stash does not apply to untracked files.
20150826 20:50:05< shadowm> Okay, I guess it can, but it's not the default.
20150826 20:50:56< Elvish_Hunter> zookeeper: just to be sure, are you using the GitHub app, or the official Git client?
20150826 20:51:12< shadowm> Neither, IIRC it's TortoiseGit.
20150826 20:51:17< zookeeper> yes
20150826 20:51:29< zookeeper> obviously you can't do anything fancy like stashing with the github app
20150826 20:51:51< zookeeper> (unless they recently added it, which i doubt)
20150826 20:52:44< shadowm> gfgtdf, zookeeper: Why not top-level [event] instead of [campaign][event]?
20150826 20:53:11< zookeeper> ok, well, i guess i don't need to stash the untracked files because there's none of my stuff in those
20150826 20:53:14< zookeeper> maybe
20150826 20:53:15< shadowm> (Top-level as in part of the same root node as [scenario], [multiplayer], [binary_patH], [units], etc.)
20150826 20:54:28< shadowm> gfgtdf, zookeeper: I'd prefer if [campaign]'s purpose was restricted to that of describing a campaign's configuration and we didn't encourage more unwieldy #ifdefs.
20150826 20:55:21< shadowm> The past tense of 'to send' is 'sent', btw.
20150826 20:56:57< zookeeper> shadowm, sure, sounds reasonable. unless there's some pitfall with the top-level thing that i can't think of... maybe UMC authors being careless, putting them outside their #ifdef's and polluting everything
20150826 20:57:28-!- oldlaptop [~quassel@50-107-94-87.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 244 seconds]
20150826 20:59:22-!- oldlaptop [~quassel@50-107-94-87.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150826 21:02:18-!- c74d3 [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20150826 21:04:54-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20150826 21:10:02-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Read error: Connection reset by peer]
20150826 21:11:12< shadowm> Not sure if you forgot that the plan is to change how add-ons are loaded so that players can disable broken add-ons without removing them.
20150826 21:12:07< shadowm> Since the add-on type will be guaranteed to be known by the client, later we can also choose to not load MP add-ons in SP and viceversa.
20150826 21:13:14< zookeeper> i don't see how that relates
20150826 21:13:39< zookeeper> except that if players can workaround a broken add-on more easily, it's not _quite_ as bad
20150826 21:14:48-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150826 21:14:55< aeth> https://en.wikipedia.org/wiki/List_of_video_game_genres#Turn-based_strategy
20150826 21:15:00< aeth> That screenshot probably needs to be updated
20150826 21:15:16< aeth> https://upload.wikimedia.org/wikipedia/commons/3/3b/Battle_for_Wesnoth_0.8.5_chaotic_indexed.png
20150826 21:17:12< aeth> (I do not have an active Wikipedia account.)
20150826 21:20:54< gfgtdf> shadowm: hm but i think in era and modifications we do it the same way with [era] define="DEFINE" #ifdef DEFINE {era_events.cfg} #enddef [/era] (although im not sure if there currently are addons using that.)
20150826 21:21:40< zookeeper> shadowm, you sure i should try to re-apply the discarded commits first? they don't touch any files i'm interested in, but do i need to do it because of some history-related black magic?
20150826 21:21:58< zookeeper> like, the stash not applying unless i do it
20150826 21:30:02< zookeeper> oh well, i guess i'll just forget about trying to salvage it and just use my usual file-copying method to workaround the git stupidity
20150826 21:31:08< zookeeper> ...and hope i didn't actually have anything else important as local changes
20150826 21:32:40< celticminstrel> gfgtdf, zookeeper, shadowm:  isn't toplevel [lua] already supported, too?
20150826 21:33:13< celticminstrel> So does that make [campaign][lua] redundant, or is there a difference between them?
20150826 21:33:26-!- gfgtdf [~chatzilla@f054128111.adsl.alicedsl.de] has quit [Read error: Connection reset by peer]
20150826 21:34:05< shadowm> Redundant, certainly.
20150826 21:35:38< shadowm> gfgtdf: See, this is where we have these irreconcilable design differences between SP and MP mode.
20150826 21:36:28< shadowm> In SP mode, campaign content is not truly loaded unless the individual campaigns are selected. In MP, *everything* is loaded by default.
20150826 21:37:11< shadowm> So MP does need more fine-grained control than SP.
20150826 21:38:12< shadowm> zookeeper: It's not black magic. I thought you wanted to get your *own* local changes back from the mush *you* created when performing the mixed reset.
20150826 21:38:27< zookeeper> umm, no
20150826 21:38:40< zookeeper> or yes, maybe
20150826 21:38:53< shadowm> For that I suggested that procedure in order to "revert" the mixed reset in a way (really just tell Git which changes are from the commit set and which aren't).
20150826 21:39:27< celticminstrel> To undo a reset, can you just reset to the original HEAD?
20150826 21:39:52< shadowm> And you keep blaming Git when the mixed reset with local changes was your idea in the first place and Git did nothing but comply with your order.
20150826 21:40:21< shadowm> Yes, you could use reflog to find out which was the original head but that means teaching zookeeper new stuff. :p
20150826 21:40:44< zookeeper> i don't see what choice i had except to reset
20150826 21:41:06< shadowm> Stashing all local changes before resetting would've been the sane choice.
20150826 21:43:14< zookeeper> well, that's why git is stupid. there's no reason why i should have assumed that reseting the way i did would cause a huge mess
20150826 21:43:40< zookeeper> but reminds me to not actually try to test PR's, i guess
20150826 21:44:21-!- gfgtdf [~chatzilla@f054128111.adsl.alicedsl.de] has joined #wesnoth-dev
20150826 21:45:14< gfgtdf> celticminstrel: no there is no bug difference but allowing just [event] and not [lua] would mean is 3 times more c++ code in that commit
20150826 21:45:24< gfgtdf> big*
20150826 21:45:37< shadowm> Thanks zookepeer, that mentality is clearly what the project needs right now.
20150826 21:45:47< shadowm> (Not.)
20150826 21:47:22< celticminstrel> Weird.
20150826 21:47:37< celticminstrel> About git, it is a little confusing, admittedly.
20150826 21:48:08< celticminstrel> I only discovered reflog a few days ago when I accidentally lost a commit during rebase somehow.
20150826 21:48:18< zookeeper> shadowm, i know
20150826 21:48:23< Elvish_Hunter> zookeeper: just to be on the safe side, I have two clones, and do all my experiments only on one
20150826 21:48:45< celticminstrel> I'd personally recommend testing PRs on a different branch than any work you're doing.
20150826 21:49:46< shadowm> ^
20150826 21:50:17< zookeeper> shadowm, to be fair, this only happened because i didn't realize that applying the .patch i got from github actually creates commits instead of local changes, if i can do the latter with the other way i didn't try then it's all good
20150826 21:50:20< shadowm> And stashing your local changes before pulling or switching branches, always.
20150826 21:50:21< celticminstrel> Though you could do what Elvish_Hunter said too. That allows two checkouts at once, which is occasionally useful...
20150826 21:50:37< zookeeper> ...well, except the fact that git has trouble dealing with local changes which is always incredibly frustrating
20150826 21:50:59< celticminstrel> I'v actually run into some weird rebase errors in the Wesnoth repo.
20150826 21:51:02< celticminstrel> ^I've
20150826 21:51:29< shadowm> You mean with the changelog? That's par for the course.
20150826 21:51:46< celticminstrel> It says "Your local changes to the following files would be overwritten by merge", but the files listed were part of the patch being applied and the working copy was clean.
20150826 21:52:07< celticminstrel> Yeah, merge errors in the changelog don't surprise me.
20150826 21:52:12< celticminstrel> That's expected.
20150826 21:52:24-!- mjs-de [~mjs-de@x4db580a5.dyn.telefonica.de] has quit [Remote host closed the connection]
20150826 21:52:33< zookeeper> in any file
20150826 21:52:41< shadowm> It happened with SVN too and this is why many projects forgo changelog files in favor of automatically generated info built from the VCS logs, but guess what, nobody here knows how to write useful commit messages.
20150826 21:52:50< celticminstrel> Heh.
20150826 21:53:03< celticminstrel> I'm not talking about the changelog here.
20150826 21:53:08< zookeeper> svn knew how to keep your local changes while pulling updates, git doesn't
20150826 21:53:10< celticminstrel> It happened with random source files.
20150826 21:53:27< BeoXTC_> you have to know how to
20150826 21:53:30< celticminstrel> git does, sort of.
20150826 21:53:34< BeoXTC_> you should use git stash
20150826 21:53:36< BeoXTC_> and git stash apply
20150826 21:53:45< zookeeper> yes, that's why git is stupid
20150826 21:53:46< celticminstrel> As long as the files changed have not been touched by what you're pulling, you don't need to stash first.
20150826 21:54:18< celticminstrel> But from what I recall, svn is a bit more intuitive, at least; git is inconsistent and confusing.
20150826 21:54:30< BeoXTC_> and for the changelog part: at our company my team is using meaningful commitmessages
20150826 21:54:35< BeoXTC_> so we can generate a changelog out of them
20150826 21:54:37< celticminstrel> Fortunately, the Internet knows everything, so if I'm stuck I just have to google.
20150826 21:55:03< BeoXTC_> we are using the same script like angular: https://github.com/angular/angular.js/blob/master/CHANGELOG.md
20150826 21:56:09< celticminstrel> Though auto-generating a changelog from commit history sounds nice, it kinda implies you need exactly one commit per feature/bugfix, which in my opinion isn't always feasible (some things really demand multiple commits in my opinion).
20150826 21:56:53< BeoXTC_> therefor you have pull requests or feature branches? depends on the feature though
20150826 21:57:23< BeoXTC_> we are using multiple commit messages for the features and it works fine enough for us to communicate what has changed in the code and functionality
20150826 21:57:43< celticminstrel> Well, the question is how the script knows which commit messages need to be included in the changelog, right?
20150826 21:58:24< BeoXTC_> there is a script, it ommits WIP: prefixed commits
20150826 21:58:42< BeoXTC_> it ommits style, test and refactor commits
20150826 21:59:10< BeoXTC_> in the end you are generating the changelog and need to fix only view lines and viola, you have the changes
20150826 21:59:17< BeoXTC_> what helped us most was this document: https://docs.google.com/document/d/1QrDFcIiPjSLDn3EL15IJygNPiHORgU1_OOAqWjiDU5Y/edit#
20150826 22:00:02< celticminstrel> I usually generate a changelog manually from the commit messages. It works because there aren't usually an enormous number of commits to go through, though; for Wesnoth that isn't feasible.
20150826 22:00:59< celticminstrel> ...okay, I want to get a vconfig from a config.
20150826 22:06:03< shadowm> gfgtdf: wesnoth.random is still missing documentation.
20150826 22:09:10-!- klemvor [~klemvor@adsl-ull-169-103.40-151.net24.it] has quit [Remote host closed the connection]
20150826 22:10:52< celticminstrel> I guess that won't work after all. :/
20150826 22:11:13< celticminstrel> I guess I can call utils::interpolate_blah_blah_blah though.
20150826 22:11:26< shadowm> celticminstrel: What are you trying to do?
20150826 22:12:52< celticminstrel> A way to patch resistances to custom damage types into predefined movetypes. I'd like to be able to set it to stuff like "$($arcane - 10)".
20150826 22:13:11< shadowm> That's syntax abuse.
20150826 22:13:36< celticminstrel> I disagree, but do you have any other suggestions?
20150826 22:14:10< shadowm> In WML land interpolation is only ever used with WML variables and it should remain this way for the sake of clarity.
20150826 22:14:37< shadowm> Unless you are going to treat $arcane as a real WML variable that exists which WML and Lua code can evaluate as such.
20150826 22:15:33< shadowm> One way to go about it, which GUI2 uses but is decidedly non-trivial, is to use the formula engine directly.
20150826 22:15:43< celticminstrel> Well, this is a completely different context, it's currently a new tag under [units]. I don't think it would make sense for it to be a real WML variable like that. If anything it's like an auto-stored variable.
20150826 22:16:03< shadowm> WML variables do not exist outside events.
20150826 22:16:13< celticminstrel> Using the formula engine directly means I'd instead write it as "arcane - 10", right?
20150826 22:16:41< shadowm> Not sure if it'd need parentheses but yes.
20150826 22:17:21< celticminstrel> Assuming it's formula.hpp, it seems to be undocumented.
20150826 22:17:44< shadowm> It probably is, yes.
20150826 22:24:24-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has left #wesnoth-dev ["Ciao!"]
20150826 22:29:29< celticminstrel> Hmm, tformula is documented...
20150826 22:31:35-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20150826 22:36:51-!- gfgtdf [~chatzilla@f054128111.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 40.0.2/20150812163655]]
20150826 22:40:49-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20150826 22:44:12-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150826 22:47:51-!- BeoXTC_ [3eb2b12a@gateway/web/cgi-irc/kiwiirc.com/ip.62.178.177.42] has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
20150826 22:52:21< celticminstrel> Well, now to test it.
20150826 22:59:13-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has quit [Remote host closed the connection]
20150826 23:12:03-!- irker607 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150826 23:20:23< celticminstrel> Apparently it doesn't work.
20150826 23:21:19< celticminstrel> ...oh. Stupid mistake.
20150826 23:34:44-!- iceiceice [~chris@23.31.228.41] has joined #wesnoth-dev
20150826 23:34:44-!- iceiceice [~chris@23.31.228.41] has quit [Changing host]
20150826 23:34:44-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150826 23:43:53-!- joet [~joet@host86-163-217-248.range86-163.btcentralplus.com] has joined #wesnoth-dev
20150826 23:44:49< celticminstrel> Works! I think.
20150826 23:45:45< celticminstrel> -20% / 2 yields -10%, which seems to make sense, but internally it's really 120 / 2 which should be 60, which would be +40%, right?
20150826 23:55:14-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150826 23:56:30-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 255 seconds]
20150826 23:58:12-!- pydsigner [~pydsigner@unaffiliated/pydsigner] has joined #wesnoth-dev
--- Log closed Thu Aug 27 00:00:13 2015