--- Log opened Fri Aug 14 00:00:08 2015
20150814 00:00:11< shadowm> vultraz, Elvish_Hunter: It's the built-in validator that's only used by GUI2 atm.
20150814 00:00:22< shadowm> (Context: http://forums.wesnoth.org/viewtopic.php?p=588480#p588480 )
20150814 00:00:50-!- travis-ci [~travis-ci@ec2-54-197-168-147.compute-1.amazonaws.com] has joined #wesnoth-dev
20150814 00:00:51< travis-ci> wesnoth/wesnoth#7032 (master - 730d837 : Charles Dang): The build failed.
20150814 00:00:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/75526117
20150814 00:00:51-!- travis-ci [~travis-ci@ec2-54-197-168-147.compute-1.amazonaws.com] has left #wesnoth-dev []
20150814 00:02:19< vultraz> :|
20150814 00:02:20< vultraz> :|
20150814 00:02:22< vultraz> :|
20150814 00:08:13-!- ancestral [~ancestral@164.sub-70-197-233.myvzw.com] has joined #wesnoth-dev
20150814 00:09:20-!- framling [~user@c-50-186-43-50.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20150814 00:11:22-!- ancestral [~ancestral@164.sub-70-197-233.myvzw.com] has quit [Client Quit]
20150814 00:28:34-!- oldlaptop_ [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150814 00:30:22-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 240 seconds]
20150814 00:30:52-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150814 00:39:42-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150814 01:07:23< ancestral> celticminstrel: Ah I see what I did
20150814 01:10:10< celticminstrel> So it works now?
20150814 01:12:51-!- oldlaptop_ is now known as oldlaptop
20150814 01:18:24-!- framling [~user@c-50-186-43-50.hsd1.or.comcast.net] has joined #wesnoth-dev
20150814 01:20:11-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150814 01:29:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150814 01:34:18< celticminstrel> I have now implemented a confirmation dialog on Cmd-Q. (Presumiably it applies to Alt-F4 as well.)
20150814 01:37:20< vultraz> celticminstrel: for quitting?
20150814 01:37:25< celticminstrel> Yes.
20150814 01:37:42< vultraz> As the mac equivalent of Ctrl Q?
20150814 01:37:50< vultraz> Because Ctrl Q brings up a quit confirmation
20150814 01:38:07< celticminstrel> Actually I think the Mac equivalent of Ctrl-Q is (weirdly) Cmd-W.
20150814 01:38:13< celticminstrel> Unless Ctrl-W is also a thing?
20150814 01:39:19< vultraz> nope
20150814 01:39:47< celticminstrel> The Quit Game option in the menu (which you recently changed to Quit to Titlescreen) is Cmd-W.
20150814 01:40:03< celticminstrel> Cmd-Q used to quit to desktop without confirmation.
20150814 01:40:15< celticminstrel> (I imagine Alt-F4 probably did the same in Windows.)
20150814 01:40:28< shadowm> Alt+F4 does that on Linux too.
20150814 01:40:46< celticminstrel> I assume that depends on your DE. :P
20150814 01:40:59< shadowm> So, what happens if you click on the window close button now?
20150814 01:41:15< celticminstrel> Same thing. Confirmation dialog. (Just confirmed that moments before you asked.)
20150814 01:41:57< shadowm> Then Alt-F4/Cmd+Q are handled by the window manager as I suspected?
20150814 01:42:17< celticminstrel> Um, SDL handles them as a special event type.
20150814 01:42:20< shadowm> I.e. Wesnoth just receives the window close event?
20150814 01:42:39< celticminstrel> Yeah.
20150814 01:42:43< celticminstrel> The quit event.
20150814 01:43:52< celticminstrel> It seems like I might've broken something, because Cmd-Q doesn't work on the titlescreen now. :(
20150814 01:44:20< shadowm> Did it before? Alt+F4 only works in windowed mode for me.
20150814 01:44:30< celticminstrel> It does on 1.12, at least.
20150814 01:44:42< celticminstrel> I'm always in windowed mode.
20150814 01:46:03< celticminstrel> I'll temporarily revert my change to see if I caused it.
20150814 02:03:13< vultraz> Ctrl Q just quits to titlescreen for me
20150814 02:03:20< vultraz> Or to desktop if from the titlescreen
20150814 02:03:57< vultraz> (testing 1.12)
20150814 02:04:32< vultraz> Same in 1.13
20150814 02:04:38< celticminstrel> Cmd-Q unconditionally quits to desktop from anywhere.
20150814 02:04:52< vultraz> celticminstrel: if you're playing with hotkeys , perhaps you can also make Esc bring up the quit confirmation dialog in the editor?
20150814 02:04:57< vultraz> I can't figure out how
20150814 02:05:12< celticminstrel> This actually has nothing to do with the hotkeys system.
20150814 02:05:18< celticminstrel> But I suppose I could look into that.
20150814 02:07:07< celticminstrel> I suspect Cmd-Q may be hard-coded into SDLmain.m or something. Maybe I'll check if that's true.
20150814 02:08:25< celticminstrel> It could also be something in the nib, but in my estimation that's unlikely since the Quit menuitem has no effect (the one in the OS menubar, not Wesnoth's menubar).
20150814 02:08:45< shadowm> vultraz: Ctrl+Q and Cmd+Q are not equivalent, so you don't want to bring up Ctrl+Q.
20150814 02:09:19< celticminstrel> It's also possible that SDL itself handles Cmd-Q specially, but that would be a bit surprising if true.
20150814 02:09:33< vultraz> ah, I see
20150814 02:09:43 * vultraz always figured ctrl = cmd on os x
20150814 02:09:51< celticminstrel> That's usually the case, yes.
20150814 02:10:23< vultraz> I mean, I have used a mac so that's how things always worked
20150814 02:10:30< vultraz> Didn't realize there was a more subtle distinction
20150814 02:10:37< celticminstrel> I'm guessing it's different here because for whatever reason Cmd-Q bypasses the normal event system.
20150814 02:10:47< celticminstrel> ie, doesn't raise keyboard events
20150814 02:11:07< celticminstrel> I would say it probably should be equivalent, honestly.
20150814 02:11:49< celticminstrel> Actually Ctrl-Q quitting to titlescreen seems weird to me.
20150814 02:12:47< vultraz> I don't know of any game that allows a direct quit to desktop from within the game
20150814 02:12:57< vultraz> Well, Bioshock does
20150814 02:13:04< celticminstrel> And Wesnoth does.
20150814 02:13:10< celticminstrel> On the Mac, it does.
20150814 02:13:15< vultraz> But any one I've seen uses Esc to get to the prompt
20150814 02:13:19< vultraz> Including wesnoth
20150814 02:13:21< celticminstrel> Cmd-Q quits Wesnoth without any confirmation.
20150814 02:13:42< celticminstrel> That's why I wanted to change this, because it's quite easy to accidentally hit Cmd-Q.
20150814 02:14:16< celticminstrel> (For example, if you meant to hit Cmd-W, Cmd-A, or Cmd-S - the first of those is, in Wesnoth, the shortcut for Quit to Titlescreen.)
20150814 02:14:46< celticminstrel> Okay, my change to the events manage did indeed cause it to not work from the titlescreen.
20150814 02:16:23< celticminstrel> It's because of GUI2.
20150814 02:16:58< vultraz> *collective groans from audience*
20150814 02:18:03< celticminstrel> For now I'm just going to throw CVideo::quit(), which means same behaviour as before.
20150814 02:18:55< celticminstrel> I'll also poke around to see if there might be any other things affected.
20150814 02:19:38< celticminstrel> Actually, CVideo::quit() is probably not at all what you'd want from pressing Cmd-Q in a GUI2 dialog. (Unless it happens to be the titlescreen.)
20150814 02:21:22< celticminstrel> So I guess I should add a new event handler whatever to GUI2 which the titlescreen hooks into to quit.
20150814 02:21:52< celticminstrel> Ah, but on the other hand, what would you expect if clicking the close button on the window decorations while a GUI2 dialog is up?
20150814 02:22:12< shadowm> A prompt?
20150814 02:22:17< vultraz> A prompt
20150814 02:22:29< vultraz> You can have more than 1 gui2 dialogs up at once
20150814 02:22:33< celticminstrel> You might be in the middle of a game, after all...
20150814 02:23:13< celticminstrel> I just hope I don't need to modify every single GUI2 dialog.
20150814 02:23:46< vultraz> p_P
20150814 02:24:09< shadowm> Of course not.
20150814 02:25:07< shadowm> GUI2 may be a bit crummy at times but it still exists within the realm of logic and reason.
20150814 02:34:06-!- StandYourGround [~Adium@2602:306:83db:de50:2831:c96b:7aa2:e377] has joined #wesnoth-dev
20150814 02:34:50< StandYourGround> Just dropping in to say I haven't forgotten about attempting a story revision for Sceptre of Fire
20150814 02:35:03< StandYourGround> I'm working really long days, but I have made progress in the writing
20150814 02:36:51-!- c74d is now known as Guest68804
20150814 02:37:05-!- Guest68804 [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Ping timeout: 245 seconds]
20150814 02:37:20< celticminstrel> I assume doofus-0 ~= StandYourGround?
20150814 02:37:23< celticminstrel> ^10
20150814 02:37:40< StandYourGround> I am not doofus-0.
20150814 02:37:43< StandYourGround> I have no alt accounts
20150814 02:37:53< celticminstrel> It's doofus-10, I typoed.
20150814 02:37:56< StandYourGround> k
20150814 02:37:58-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20150814 02:38:49< StandYourGround> So basically, I'm planning to address the worst plot holes and let the rest stay in in a more believable way
20150814 02:39:07< celticminstrel> I don't think doofus-10 was doing writing stuff, but not sure.
20150814 02:39:33< StandYourGround> The story will be narrated from the perspective of a visitor to Hamel's court telling his version of the story for entertainment
20150814 02:39:59< StandYourGround> and most of the weird stuff that happens in the story (those elves make no sense) will be the result of the dwarves arguing over details
20150814 02:40:13< vultraz> Sounds good to me
20150814 02:40:20< celticminstrel> What's these elves? (I haven't played it)
20150814 02:40:41< StandYourGround> The one thing I intend to completely change is removing Krawg the stupid talking gryphon for good
20150814 02:40:49< celticminstrel> Oh, I was wrong, it's doofus-01, not doofus-10.
20150814 02:41:24< StandYourGround> and killing Thursagan, leaving just his ghost as the new sidekick, having guarded his hammer waiting for someone worthy to train to use it
20150814 02:41:39< celticminstrel> Yeah, looks like he mostly did more mechanical revisions, though I see some story stuff.
20150814 02:41:50< celticminstrel> I have no idea if that's going to be merged.
20150814 02:42:16< StandYourGround> Krawg is like the worst sidekick and worst plot hold ever
20150814 02:42:36< StandYourGround> seriously, if he could fly out of the volcano at the end, why not take at least the Sceptre along, if not a few dwarves as well?
20150814 02:43:53< StandYourGround> but instead the sidekick will be Thursagan's ghost busily at work at his final great creation, then sadly unable to carry it with him as he escaped the volcano.
20150814 02:44:06< StandYourGround> because he's all incorporeal, and can't actually carry a physical object
20150814 02:44:14< shadowm> :\
20150814 02:44:51< StandYourGround> but rumor has it he's still watching over the Sceptre…  
20150814 02:45:00< StandYourGround> (it would be cool if that happened in HtTT someday)
20150814 02:45:06< celticminstrel> Incidentally, https://github.com/wesnoth/wesnoth/pull/436
20150814 02:46:37< StandYourGround> there is no purpose for gryphons or gryphon riders in a game where most of the action happens on cave terrain
20150814 02:46:44< StandYourGround> campaign 
20150814 02:47:11< StandYourGround> and attempting to transliterate a grypon that's about as literate as a toddler looks cheap and silly 
20150814 02:48:13< StandYourGround> I've nothing against a mainline campaign with humor…  just less stupid humor would be more professional :P
20150814 02:48:34< StandYourGround> and there will be humor.  :)
20150814 02:49:14< StandYourGround> AI sides changing factions or alliances as a result of increasingly drunk dwarves arguing over details in the story they don't like…
20150814 02:49:29< StandYourGround> (the narrator is from the Shorbear Clan, so he has some biases)
20150814 02:51:11-!- Xara [~Yangyf@2401:de00:1:6:6449:f953:1854:6a14] has joined #wesnoth-dev
20150814 02:52:16< StandYourGround> Well…  I intend to finish more of it very soon..  I'm following the format of the Dialogue page from the wiki
20150814 02:52:39-!- Brandon [~Brandon@ip72-194-95-216.oc.oc.cox.net] has joined #wesnoth-dev
20150814 02:52:48< StandYourGround> So I have all the death dialogue and the intro to the first scenario finished, and I'm working through the dialogue
20150814 02:52:57-!- Brandon is now known as Guest66574
20150814 02:54:18< StandYourGround> I do have a question…  should this rewrite ever eventually lead to a mainline change, how difficult would it be to "fade out" the game (darken the screen or something) to transition from the game back to the narrative…  I'm planning on having the gameplay "interrupted" occasionally as the dwarves in Hamel's court interrupt the storyteller
20150814 02:54:35< celticminstrel> I believe that's relatively easy.
20150814 02:55:23< StandYourGround> I think all the narrator/audience dialogue could be handled the standard way, and there shouldn't ever be a need to create a map to show the actual feast in Hamel's court
20150814 02:55:32< celticminstrel> Trying to find something.
20150814 02:55:55< celticminstrel> Ah, there it is.
20150814 02:55:59< celticminstrel> {FADE_TO_BLACK}
20150814 02:56:13< StandYourGround> it wouldn't be necessary to do a complete fade...
20150814 02:56:19< celticminstrel> If you want something more complicated you could use [color_adjust}.
20150814 02:56:26< celticminstrel> ^[color_adjust]
20150814 02:57:00< celticminstrel> {FADE_TO_BLACK} is implemented as a series of [color_adjust]...[/color_adjust][delay]...[/delay][redraw][/redraw].
20150814 02:57:07< vultraz> ^
20150814 02:57:09< StandYourGround> there is one slight timeline placement change my rewrite will have
20150814 02:57:14< celticminstrel> It's in data/core/interface-utils.cfg.
20150814 02:57:25< StandYourGround> as it is now, the first scenario appears to be happening after Haldric I has died
20150814 02:57:32< StandYourGround> and his son Haldric II is the king of Wesnoth
20150814 02:57:43< celticminstrel> I guess if you want something more complicated it'd be simpler to actually use {FADE_STEP_RGB}.
20150814 02:58:21< StandYourGround> I want to change it so that Haldric I is still alive and very near the end of his reign, very corrupted by the Ruby of Fire, and his son has taken it against his father's wishes to have it made into the Sceptre
20150814 02:58:39< StandYourGround> fearing mostly for what the Ruby could do to him in the future
20150814 02:59:15< StandYourGround> so in the first scenario, the attackers start out as humans, presumably mercenaries hired by Haldric I to reclaim the Ruby without using the Wesnoth army against his son
20150814 02:59:46< StandYourGround> but this doesn't sit well with the dwarves listening, so after some debate, they agree that the attackers might as well be elves instead
20150814 02:59:59< StandYourGround> so the human enemies on the map are simply replaced with elves
20150814 03:00:24< StandYourGround> this change has effects later, as the Shorbear narrator doesn't like the idea of his clan allying with the elves
20150814 03:00:38< StandYourGround> so he tries to change the story to somehow make the Elves on Rugnur's side
20150814 03:00:47< StandYourGround> which confuses even Rugnur
20150814 03:00:55< StandYourGround> (little fourth wall tampering here)
20150814 03:01:16< StandYourGround> after a few turns and narrator arguments, the narrator gives in after being sufficiently bribed with extra ale
20150814 03:02:04< StandYourGround> and the elves are switched back and forth from ally/enemy/xenophobic to finally allying with the Shorbear dwarves as they do in the current version of SoF
20150814 03:02:06< celticminstrel> Pffft.
20150814 03:02:19< celticminstrel> I dunno if I like that, but it is at least hilarious.
20150814 03:03:19< StandYourGround> well, when I finally have it written enough to post on the Writer's forum, we can debate it :D
20150814 03:03:29< StandYourGround> I don't have great expectations to see it mainline immediately
20150814 03:03:40< StandYourGround> but if after some revisions and if the community supports it...
20150814 03:03:53< StandYourGround> who knows
20150814 03:06:09-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has joined #wesnoth-dev
20150814 03:06:40-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150814 03:11:28< StandYourGround> Also, for the Elves & Shorbear vs. Rugnur scenario, the Elf Leader will have a bit of Deadpool-like awareness of the tampering with his side
20150814 03:12:10< StandYourGround> he wants to say his epic speech declaring war on his enemy, and has to start over everytime his side is switched.
20150814 03:12:50< StandYourGround> "Okay, let's try this again…  prepare to die, foul dwarves who trespass our lands and steal what is rightfully ours!"
20150814 03:13:04< StandYourGround> "Okay… never mind.  All of you foul dwarves must die..."
20150814 03:14:14< StandYourGround> "what, again?  Okay, fine.  Um…  Dwarves of the Shorbear Clan!  These thieving dwarves that attack you have stolen from us, and we only require what they have taken.  We will aid you in your defense."
20150814 03:15:00< StandYourGround> something like that
20150814 03:15:05< StandYourGround> still a work in progress
20150814 03:15:15< StandYourGround> I'll be back when I have more :D
20150814 03:20:47-!- StandYourGround [~Adium@2602:306:83db:de50:2831:c96b:7aa2:e377] has quit [Quit: Leaving.]
20150814 03:29:49-!- Guest66574 [~Brandon@ip72-194-95-216.oc.oc.cox.net] has quit [Quit: Leaving]
20150814 03:29:51-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Disconnected by services]
20150814 03:32:02-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150814 03:40:09-!- Appleman1234 [~Appleman1@KD036009070013.au-net.ne.jp] has quit [Ping timeout: 246 seconds]
20150814 03:50:00-!- Xara [~Yangyf@2401:de00:1:6:6449:f953:1854:6a14] has quit [Read error: Connection reset by peer]
20150814 03:50:13-!- irker053 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150814 03:52:11< celticminstrel> So I currently have GUI2 quitting to desktop upon receiving a quit event. I'm not quite sure what I would need to do to allow dialogs to specify different behaviour. (Actually the default behaviour should probably be something like "quit to desktop after a confirmation".)
20150814 03:54:38< Aginor> celticminstrel: from a usability perspective, it's probably better to offer to save the game (if not saved), and offer "ok, cancel" instead of asking "do you really want to quit?"
20150814 03:55:14< celticminstrel> Saving the game in the middle of an event seems like a bad idea.
20150814 03:55:41< Aginor> how is it normally done?
20150814 03:55:50-!- N4tr0n [~user@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150814 03:55:57< Aginor> I expect there to be a mouse-clicked event for the "ok" button if nothing else
20150814 03:56:27-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has quit [Quit: "The highest possible stage in moral culture is when we recognize that we ought to control our thoughts." - Charles Darwin]
20150814 03:56:41< celticminstrel> Unrelated: It would be nice if there were a view on units.wesnoth.org with all optional advancements enabled.
20150814 03:56:55< Aginor> but I suspect that the better course of action would be to handle the event that triggers the dialog, the dialog itself then terminates the program
20150814 03:57:06-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has joined #wesnoth-dev
20150814 03:59:19< Aginor> clarifiaction: I suspect that the better course of action would be to handle the quit event. The event handler triggers the dialog. If the dialog returns a confirmation of the actions, quit the game
20150814 03:59:45< Aginor> I'm not sure how the high-level events are handled, but I would image that it can still be done within the event-loop
20150814 03:59:54< celticminstrel> I'm not sure.
20150814 04:00:38< Aginor> hmmm
20150814 04:00:48< celticminstrel> I'm going to commit the unconditional quit because it restores previous behaviour, but I think this does need work.
20150814 04:00:53< Aginor> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=462794 hasn't been properly fixed
20150814 04:01:36< Aginor> the little ugly fix in display.cpp:2243 needs to be brought into other places that call SDL_BlitSurface
20150814 04:02:01< Aginor> s/SDL_BlitSuface/SDL_Blitsurface/
20150814 04:05:20< Aginor> I'll have a pull request ready in a few minutes
20150814 04:09:22< celticminstrel> It could be nice to have additional hooks into the markov chain generator.
20150814 04:10:37< celticminstrel> For example, something in [set_variable] (or a whole new WML tag), and support for custom namelists in specific unit types (overriding the race names).
20150814 04:18:43-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 244 seconds]
20150814 04:23:51-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20150814 04:37:54-!- Appleman1234 [~Appleman1@KD106179115246.au-net.ne.jp] has joined #wesnoth-dev
20150814 04:41:32-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150814 04:41:32-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150814 04:41:32-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150814 04:42:14-!- higgins [~higgins@105.ip-167-114-152.net] has quit [Quit: Leaving]
20150814 04:47:10-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150814 04:48:12-!- higgins [~higgins@105.ip-167-114-152.net] has joined #wesnoth-dev
20150814 05:01:33-!- Kwandulin [~Miranda@p5B0080CB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150814 05:31:51< shadowm> ancestral: I'm going to ask you a difficult question.
20150814 05:32:24< shadowm> Well, I'm not sure if celticminstrel can answer this question too.
20150814 05:32:42< shadowm> ancestral: Which one does the Xcode project use atm? filesystem_boost.cpp, or filesystem.cpp?
20150814 05:33:09< ancestral> celticminstrel: I’m trying to build the project right now, FYI. Thanks for your patience
20150814 05:33:19< ancestral> I passed out earlier tonight
20150814 05:33:23< ancestral> shadowm: Hmm
20150814 05:33:51< celticminstrel> shadowm: The former.
20150814 05:34:01< shadowm> Okay, then I'm killing off filesystem.cpp.
20150814 05:44:26-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has quit [Ping timeout: 260 seconds]
20150814 05:57:04< shadowm> We have a direct dependency on shell32.dll.
20150814 05:58:23< shadowm> Oh for the love of god Microsoft.
20150814 05:59:08< shadowm> SHGetSpecialFolderPath -> ShGetFolderPath -> SHGetKnownFolderPath
20150814 06:00:16< shadowm> I guess I'll figure out later how to avoid this runtime linking nonsense that makes the code smell like fish.
20150814 06:03:34-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150814 06:07:01-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150814 06:09:59-!- Xara [~Yangyf@2401:de00:1:6:90b0:1b96:cbcd:3951] has joined #wesnoth-dev
20150814 06:13:31-!- markus_ [~mjs-de@x4db68cce.dyn.telefonica.de] has joined #wesnoth-dev
20150814 06:13:50< shadowm> loonycyborg, aquileia, vultraz: Re bug #23753, what do you feel the new behavior should be if a relative path is specified in the command line?
20150814 06:14:52< shadowm> *Relative* (e.g. "wesnothcrap") as opposed to *absolute* (e.g. "C:\wesnothcrap").
20150814 06:15:43< shadowm> Should it continue to be relative to Documents\My Games?
20150814 06:16:56-!- mjs-de [~mjs-de@x4db50479.dyn.telefonica.de] has quit [Ping timeout: 250 seconds]
20150814 06:20:10< shadowm> On Linux they'd be relative to %HOME% (the Windows equivalent being %HOMEDRIVE%%HOMEPATH%).
20150814 06:20:15< shadowm> *$HOME
20150814 06:21:49< celticminstrel> The Windows equivalent to ~/Library/Application Support is %APPDATA%, so that would also be a logical place for it.
20150814 06:22:28< shadowm> Except that if someone is messing with the command line you'd think that the last thing they want is for Wesnoth to write into one of the hidden dirs.
20150814 06:22:42< celticminstrel> If a relative data/userdir path is specified from the command line, I would expect it to be taken relative to the current working directory.
20150814 06:23:10< celticminstrel> Currently the datadir path is taken relative to the Resources folder in the application package.
20150814 06:23:39< celticminstrel> It'd also be reasonably logical to take it relative to the executable (but probably harder? I dunno).
20150814 06:24:31< shadowm> No.
20150814 06:24:40< celticminstrel> If I wanted it to be relative to $HOME, I would use the ~ notation.
20150814 06:24:50< shadowm> Which does not exist in Windows.
20150814 06:24:54< celticminstrel> True.
20150814 06:26:16< shadowm> I know what the new default should be, I know what to do with absolute paths, I just don't know whether to keep the existing behavior with relative paths or change them to make them relative to CWD, the exe dir (which is currently the same as CWD *but I'm going to change this later too*), or the user profile dir.
20150814 06:26:30< shadowm> The very last option would be congruous with Linux practice.
20150814 06:27:32-!- genbattle_ [~genbattle@115-188-147-250.jetstream.xtra.co.nz] has joined #wesnoth-dev
20150814 06:27:51< celticminstrel> Current behaviour is... relative to the default location?
20150814 06:28:06< shadowm> No.
20150814 06:28:29< shadowm> loonycyborg, aquileia, vultraz: I'm going to rant a little here, ignore this.
20150814 06:29:10< shadowm> celticminstrel: The current behavior is to make them relative to Documents\My Games. This is used by the installer to make Wesnoth use Documents\My Games\WesnothX.Y as its user data and config dir.
20150814 06:29:28< celticminstrel> What's the current behaviour on Mac/Linux, then?
20150814 06:29:29< shadowm> The whole point of me dealing with bug #23753 is to get rid of the current behavior.
20150814 06:29:38-!- framling [~user@c-50-186-43-50.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20150814 06:30:03< shadowm> The current behavior in OS X is "I don't know and I don't want to know because I'd prefer to deal with one issue at a time, and only on platforms I know".
20150814 06:30:28< shadowm> The current behavior on Linux and other X11 platforms is to use paths relative to %HOME%.
20150814 06:30:34< shadowm> Argh, $HOME. Why do I keep doing that.
20150814 06:30:57< celticminstrel> Could be the same on OSX.
20150814 06:31:03< shadowm> Note that I'm specifically talking of the behavior of --config-dir when given a relative path. I'm not describing the default behavior of Wesnoth when --config-dir is not specified at all!
20150814 06:31:10< celticminstrel> Right.
20150814 06:31:17< shadowm> That is a wholly different can of worms.
20150814 06:31:34< celticminstrel> Relative to $HOME isn't much different from relative to Documents\My Games.
20150814 06:31:55< shadowm> On Linux/X11, what happens is that Wesnoth is normally compiled without a default user config+data dir preference.
20150814 06:32:12-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 264 seconds]
20150814 06:32:34< shadowm> This means Wesnoth will use the XDG layout for this, splitting the layout into two separate dirs, one for config, and another for data.
20150814 06:32:51< shadowm> That is, when --config-dir or --userconfig-dir or --userdata-dir are not provided in the command line.
20150814 06:33:31< shadowm> So, .config/wesnoth for user config, and .local/share/wesnoth/X.Y for user data.
20150814 06:34:17< shadowm> IF Wesnoth is compiled with a default preference, the default behavior will be to use that path relative to $HOME even if no switch is provided.
20150814 06:34:37< shadowm> This is the classic layout and it's what I've continued to use all these years.
20150814 06:35:09< shadowm> That makes the handling of relative paths in --*-dir switches equivalent to changing the compiled-in preference.
20150814 06:35:46< shadowm> However, applying the same logic on Windows is not very practical.
20150814 06:36:46< shadowm> The new default when no --*-dir switches are provided will be to use Documents\My Games\WesnothX.Y (so, like previously with a relative path in --config-dir).
20150814 06:37:16< shadowm> The question is whether to keep handling --config-dir the same way or make any passed paths relative to something else.
20150814 06:37:57< shadowm> Backwards compatibility is not the issue here because this change *will* break things for people who were using the thrice-damned legacy .\userdata layout.
20150814 06:38:57< shadowm> (And I don't care because these are development versions and the breakage is only relevant to a very specific subset of people with a misguided notion of what a certain now-removed installer option does.)
20150814 06:44:48-!- genbattle_ [~genbattle@115-188-147-250.jetstream.xtra.co.nz] has quit [Ping timeout: 264 seconds]
20150814 06:45:58-!- Appleman1234 [~Appleman1@KD106179115246.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20150814 06:46:26-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150814 06:48:01-!- Xara2 [~Yangyf@2401:de00:1:6:70a6:c318:1398:5f22] has joined #wesnoth-dev
20150814 06:48:25-!- Yang_Yifan [~Yangyf@2401:de00:1:6:78f6:845a:6a8a:cee5] has joined #wesnoth-dev
20150814 06:51:02-!- [Relic] [~Relic]@2602:306:33a3:6d30:d41:83e3:7c7d:43ca] has quit [Quit: I press the magic X and all the weirdos go away!]
20150814 06:51:08-!- Xara [~Yangyf@2401:de00:1:6:90b0:1b96:cbcd:3951] has quit [Ping timeout: 244 seconds]
20150814 06:52:35-!- Xara2 [~Yangyf@2401:de00:1:6:70a6:c318:1398:5f22] has quit [Ping timeout: 250 seconds]
20150814 07:12:10< shadowm>     User configuration directory: C:\users\shadowm\My Documents\My Games\Wesnoth1.13
20150814 07:12:13< shadowm> \o/
20150814 07:20:16< ancestral> celticminstrel: Your patch is great
20150814 07:20:22< celticminstrel> Yay.
20150814 07:20:38< celticminstrel> Funny timing actually, as a few moments ago I stopped myself from poking you about it.
20150814 07:20:51< ancestral> I had to slash and burn everything
20150814 07:20:57< ancestral> I fscked up my local dir too much
20150814 07:21:23< ancestral> If I was more delicate I’d have started a local branch
20150814 07:21:32< ancestral> Then just switched branches
20150814 07:21:35< celticminstrel> On a random note I think PR449 should be given priority.
20150814 07:21:42< ancestral> 449?
20150814 07:22:07< celticminstrel> Password hashing in campaignd.
20150814 07:22:44< ancestral> Ah, that’s out of my league
20150814 07:22:57< celticminstrel> I assumed as much, it wasn't addressed at you specifically.
20150814 07:23:12< ancestral> However, the other thing, I guess I can approve that PR
20150814 07:23:19< shadowm> The campaignd maintainer is shadowm.
20150814 07:23:20< ancestral> Tomorrow so I can sleep
20150814 07:23:26< celticminstrel> On a related note, I don't like how the passphrase is in the pbl file together with important data.
20150814 07:23:34< shadowm> I'm fairly sure he's aware of the PR and is just busy with a few other things.
20150814 07:24:05< celticminstrel> Barring the passphrase, I feel that the pbl belongs in version control.
20150814 07:24:32< shadowm> That's why my proposed model moves all the data bar passphrase and email to a separate file.
20150814 07:24:40< celticminstrel> Ooh, okay then.
20150814 07:24:43< shadowm> You know, from that topic nobody has read.
20150814 07:24:59< shadowm> (Good job community, you are doing a stellar job at keeping me interested in my own proposal.)
20150814 07:25:34< celticminstrel> Good to know it's being thought of.
20150814 07:27:26< loonycyborg> shadowm: you consider moving it somewhere else than Saved Games at least on win7+?
20150814 07:27:42< loonycyborg> I mean
20150814 07:27:47< loonycyborg> My Games
20150814 07:27:50< shadowm> Oh god here we go again.
20150814 07:27:56< loonycyborg> confused
20150814 07:28:00< shadowm> loonycyborg: Relative path handling only.
20150814 07:28:29< loonycyborg> there's more present paths on newer version of windows
20150814 07:28:44< loonycyborg> *preset
20150814 07:29:00< shadowm> Nobody's said anything about Saved Games.
20150814 07:29:34< loonycyborg> But maybe change what they're relative to?
20150814 07:29:43< celticminstrel> The forum's being slow to load. I hope it's not down right when I want to look up shadowm's topic.
20150814 07:29:54< loonycyborg> to some path that is localized
20150814 07:29:54< shadowm> loonycyborg: (For the nth time) the default will be to use Documents\My Games\WesnothX.Y if *no* command line options are used. Now I want to know what to do when --config-dir/--userconfig-dir/--userdata-dir are used with a *relative* path.
20150814 07:30:17< shadowm> I can either keep the current behavior (which is *identical to the new default*) or change it.
20150814 07:31:52< loonycyborg> hmm I think it should be relative to one of windows preset paths, like documents
20150814 07:31:52< loonycyborg> or more specific one
20150814 07:33:57< loonycyborg> https://msdn.microsoft.com/en-us/library/windows/desktop/bb762188%28v=vs.85%29.aspx
20150814 07:34:18< loonycyborg> https://msdn.microsoft.com/en-us/library/windows/desktop/dd378457%28v=vs.85%29.aspx
20150814 07:34:21< shadowm> *SIGH*
20150814 07:34:33< loonycyborg> FOLDERID_SavedGames
20150814 07:34:42< shadowm> I had this stuff open in other tabs since hours ago.
20150814 07:35:01< shadowm> It's not what I want to deal with and we already decided to keep using Documents\My Games.
20150814 07:35:50< loonycyborg> there's a bug report of My Games not being translated
20150814 07:36:02< loonycyborg> FOLDERID_SavedGames would be localized by windows
20150814 07:36:13< shadowm> I am aware of this too.
20150814 07:36:50< loonycyborg> well then make me aware of your stance on this issue and how it should be fixed. If it should.
20150814 07:37:11< shadowm> I already did.
20150814 07:37:21< shadowm> <shadowm> bug #23753
20150814 07:37:31< shadowm> https://gna.org/bugs/?23753
20150814 07:39:23< loonycyborg> yes I got that and agree but
20150814 07:39:52< loonycyborg> I see no justification of using My Games rather than FOLDERID_SavedGames
20150814 07:40:21< shadowm> Firstly, Vista was already around when you introduced this layout, so I wonder what your logic was back then.
20150814 07:40:55< shadowm> Secondly, aquileia and others pointed out several AAA titles use unlocalized Documents\My Games instead of whatever Microsoft's latest recommendation is.
20150814 07:41:16< shadowm> In view of this, I declare this to be a political issue not worth of my time.
20150814 07:41:41< loonycyborg> yes, I just blindly copied from those AAA titles
20150814 07:41:56< loonycyborg> I suspect this choice is due to histerical reasons
20150814 07:41:58< shadowm> Thirdly I'd rather make all Windows versions (including Wine) use a single standard.
20150814 07:42:31< shadowm> This is both for UX reasons (what happens if you upgrade Windows? how more convoluted can we make the relevant documentation on the wiki?) and to keep the code simple.
20150814 07:42:47< loonycyborg> hmm ok
20150814 07:43:32-!- Appleman1234 [~Appleman1@KD106179122109.au-net.ne.jp] has joined #wesnoth-dev
20150814 07:43:38< shadowm> Until we officially drop support for Windows XP this means that we should not depend on a post-XP shell path or alternatively we'll have to take it on our hands to create it every time (like we already do with My Games) which is annoying.
20150814 07:44:00< shadowm> So, again, political issue. It can be deal with separately.
20150814 07:44:06< shadowm> My question to you was very specific.
20150814 07:44:13-!- Shackra [~Jorge@186.177.2.148] has quit [Ping timeout: 246 seconds]
20150814 07:44:30< shadowm> Now that relative path handling becomes the default, should relative path handling change to use something else or be a mirror of the default?
20150814 07:44:52< loonycyborg> I'm not sure myself but
20150814 07:44:59< loonycyborg> there's one reason to change it
20150814 07:45:14< loonycyborg> to remove non-localized fragment "My Games"
20150814 07:45:51< loonycyborg> so people who want to localize it could do it with relpath in --config-dir
20150814 07:46:04< loonycyborg> but on the other hand not very good idea for them to do that
20150814 07:46:20< loonycyborg> this path will override version numbers in WesnothX.Y path
20150814 07:46:23< loonycyborg> part
20150814 07:47:34< loonycyborg> another option is forget about localized path issue until we drop winxp
20150814 07:47:42-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150814 07:47:45< loonycyborg> and keep it as is
20150814 07:49:55-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150814 07:50:14< shadowm> So relative to Documents instead of Documents\My Games is what you propose.
20150814 07:50:23< loonycyborg> yes
20150814 07:50:46< shadowm> The reasoning seems extremely frivolous to me.
20150814 07:50:56< loonycyborg> I had another even more frivolous idea
20150814 07:51:19< shadowm> Particularly so because most users are not expected to learn that --config-dir even exists.
20150814 07:51:28< loonycyborg> we could move some of xdg path parts to %APPDATA%
20150814 07:51:36< loonycyborg> like cache maybe
20150814 07:51:53< shadowm> So the validity of this argument is limited to a very particular crowd who can presumably deal with gratuitous English in their filesystem anyway.
20150814 07:52:45< shadowm> XDG is an X11 land thing. It's in its very name.
20150814 07:53:16< shadowm> I know what you actually mean but that's really stretching the terminology.
20150814 07:53:52-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 272 seconds]
20150814 07:53:52-!- wedge010 is now known as wedge009
20150814 07:54:07< shadowm> If you asked me, I'm not a fan of the XDG layout and this is why even to this day I keep using the prefs dir option with SCons for my personal builds.
20150814 07:54:24< shadowm> I'd rather have everything in a single dir like we currently do in OS X and Windows and used to do by default on Linux.
20150814 07:55:08< shadowm> I also don't think Windows users (esp. WML coders) deserve adding another complication on top of the fact that they are using a crummy OS.
20150814 07:58:40< loonycyborg> yes, I just used this terminology for clarity because currently unix builds are affected by this distrinction
20150814 07:59:28< loonycyborg> but it seems to me that those xdg path conventions are influenced by windows and appdata vs documents dirs
20150814 07:59:37< loonycyborg> so that's where this idea comes from :P
20150814 07:59:42< shadowm> It's a really stupid idea.
20150814 08:00:40< shadowm> I know the purpose is to reduce the dotfile count in $HOME, but there were better ways to go about it without complicating everyone's lives.
20150814 08:05:37< shadowm> You know, it just dawned on me...
20150814 08:06:17< shadowm> If you install Wesnoth to a removable device, you *cannot* reliably use an absolute path to get the old "user config+data in install dir" behavior (a.k.a. .\userdata).
20150814 08:06:57< loonycyborg> actually yes
20150814 08:07:01< shadowm> This is because the removable volume's drive letter is likely to vary over time (especially but not exclusively if you use the same install on different systems).
20150814 08:07:55< shadowm> So perhaps the new default should be to use paths relative to the process current working dir?
20150814 08:07:59< shadowm> aquileia, vultraz: ^
20150814 08:08:25< loonycyborg> hmm
20150814 08:08:37< loonycyborg> another way would be add special option
20150814 08:08:55< loonycyborg> to treat all paths relative to cwd
20150814 08:10:14< shadowm> Or treat any path starting with . or .. especially.
20150814 08:10:41< loonycyborg> hmm that might get confusing
20150814 08:11:29< shadowm> Incidentally, --config-dir will be renamed "soon" (haven't decided to what) since it's no longer a very descriptive switch now that --user{config|data}-dir exists.
20150814 08:11:44< loonycyborg> does any software make distriction between purely relative paths and paths starting with .?
20150814 08:12:23< shadowm> No idea.
20150814 08:16:35-!- irker035 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150814 08:16:35< irker035> wesnoth: Ignacio R. Morelle wesnoth:master d3d897060120 / changelog src/filesystem.cpp: fs: Remove legacy non-Boost implementation of the filesystem API http://git.io/v3XrR
20150814 08:17:18-!- Shackra [~Jorge@186.177.2.148] has quit [Ping timeout: 272 seconds]
20150814 08:17:20< shadowm> Let's see who will be the first to complain about broken builds because they missed the memo back in November last year.
20150814 08:20:15< celticminstrel> Heh.
20150814 08:21:35< shadowm> --coonfig-dir -> --user-dir I think is the most reasonable choice.
20150814 08:21:50< shadowm> --userconfig-dir + --userdata-dir = --user-dir.
20150814 08:24:10-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150814 08:33:11-!- Yang_Yifan [~Yangyf@2401:de00:1:6:78f6:845a:6a8a:cee5] has quit [Read error: Connection reset by peer]
20150814 09:00:54-!- travis-ci [~travis-ci@ec2-54-163-75-11.compute-1.amazonaws.com] has joined #wesnoth-dev
20150814 09:00:55< travis-ci> wesnoth/wesnoth#7034 (master - d3d8970 : Ignacio R. Morelle): The build is still failing.
20150814 09:00:55< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/75567547
20150814 09:00:55-!- travis-ci [~travis-ci@ec2-54-163-75-11.compute-1.amazonaws.com] has left #wesnoth-dev []
20150814 09:03:50< Aginor> can someone please try to compile https://github.com/wesnoth/wesnoth/pull/452 on windows and OSX to make sure it's warnings free there too? Preferably with and without SDL2 enabled.
20150814 09:07:21< shadowm> Good luck getting a Windows user to compile with SDL 2. :p
20150814 09:08:46< Aginor> oh?
20150814 09:09:02< Aginor> what's the problem?
20150814 09:09:16-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150814 09:09:33< shadowm> Windows people tend to shy away rfom anything that requires changing their build environment no matter how slightly.
20150814 09:10:46< shadowm> This is because they usually have to hunt for the required libraries and headers by themselves for the lack of a simple uniform installation mechanism like `sudo apt-get install libsdl2-dev`.
20150814 09:11:07< Aginor> yeah, fair enough
20150814 09:11:50< shadowm> Our OS X packagers have the same problem. No idea about other OS X users able to compile.
20150814 09:12:21< Aginor> unforunately I want to know if my (void)(foo); way of dealing with unused variables with clang, visualstudio and minigw
20150814 09:12:29< Aginor> +works
20150814 09:13:05< shadowm> (void)foo has served us well when dealing with my code at least.
20150814 09:13:18< Aginor> ok, sounds good
20150814 09:13:56< Aginor> I kind of would like to define the good old UNUSED(foo); macro, where would be a good place for it?
20150814 09:14:19< shadowm> E.g. https://github.com/wesnoth/wesnoth/blob/master/src/desktop/open.cpp#L110
20150814 09:15:15< Aginor> then it'll hopefully be safe
20150814 09:15:18< shadowm> Aginor: There is src/global.hpp, but you might want to make sure it doesn't conflict with Lua's identical macro somewhere.
20150814 09:15:21< shadowm> src/lua/llimits.h:88:#define UNUSED(x)  ((void)(x))     /* to avoid warnings */
20150814 09:15:40< Aginor> hmm
20150814 09:15:54< shadowm> No idea if that file is visible to one or more of the files in src/scripting/.
20150814 09:16:29< Aginor> solvable by undefining previous defines for UNUSED in lua though
20150814 09:18:30< Aginor> ah, it's safe already
20150814 09:19:32< shadowm> wedge009: Re PR #450, uh, what would happen if I want to bind Ctrl+j and not Ctrl+Enter somewhere?
20150814 09:20:37< shadowm> (Ctrl+j is used by default for the objectives display right now.)
20150814 09:27:19-!- Kwandulin [~Miranda@p5B0080CB.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150814 09:28:02-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 250 seconds]
20150814 09:37:37-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 264 seconds]
20150814 09:40:30-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20150814 09:42:20-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has joined #wesnoth-dev
20150814 09:50:20-!- Xara [~Yangyf@2401:de00:1:6:78f6:845a:6a8a:cee5] has joined #wesnoth-dev
20150814 09:51:36< Aginor> hmm, someone's broken the WML tests
20150814 09:52:14-!- markus_ is now known as mjs-de
20150814 10:09:42-!- subliun [~quassel@139.216.140.254] has quit [Ping timeout: 246 seconds]
20150814 10:10:29-!- subliun [~quassel@139.216.140.254] has joined #wesnoth-dev
20150814 10:18:34-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 246 seconds]
20150814 10:26:55< zookeeper> in the editor, the custom time schedule RGB sliders behave really weirdly.
20150814 10:27:26< zookeeper> if i grab a slider and drag it in one direction, it lags behind the pointer, and when i stop moving but keep the button pressed, it starts sliding back towards 0
20150814 10:27:48< zookeeper> correction: it slides left, not towards 0
20150814 10:30:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150814 11:05:52-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20150814 11:05:55-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20150814 11:09:24-!- Kwandulin [~Miranda@p5B0080CB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150814 11:13:07-!- Grickit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20150814 11:15:14-!- Grickit is now known as Gambit
20150814 11:16:06< irker035> wesnoth: Andreas Löf wesnoth:master 46afc83f3a93 / src/ (sdl/utils.cpp sdl/utils.hpp unit_drawer.cpp video.cpp widgets/menu.cpp): Fix all compile warnings when SDL2 is enabled http://git.io/v310x
20150814 11:16:08< irker035> wesnoth: Andreas Löf wesnoth:master f0eba4b3124d / src/ (desktop/open.cpp global.hpp tests/test_config.cpp video.cpp): Add a macro to clearly mark unused variables as unused and avoid compiler warnin http://git.io/v310p
20150814 11:16:10< irker035> wesnoth: lipk wesnoth:master 0d6fdfe2cc68 / src/ (8 files in 5 dirs): Merge pull request #452 from aginor/sdl2 http://git.io/v310j
20150814 11:17:24-!- lipkab [~the_new_l@host-91-147-210-193.biatv.hu] has joined #wesnoth-dev
20150814 11:21:50< lipkab> Aginor: By tradition, we offer team membership for new contributors after their second patch.
20150814 11:22:26< lipkab> I've just merged a second pull request by you, so do you want to become a Wesnoth developer? :)
20150814 11:24:57-!- Appleman1234 [~Appleman1@KD106179122109.au-net.ne.jp] has quit [Ping timeout: 256 seconds]
20150814 11:30:05-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20150814 11:37:10-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150814 11:54:21< Aginor> lipkab: hi
20150814 11:54:48< Aginor> lipkab: it would speed things up as I get to the serious stuff :)
20150814 11:54:56< Aginor> so yes
20150814 11:57:31< Aginor> lipkab: If you have the time, I'd like to ask your input on some of my thoughts for the scrolling bug too
20150814 11:57:35-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150814 11:57:35-!- travis-ci [~travis-ci@ec2-54-144-26-151.compute-1.amazonaws.com] has joined #wesnoth-dev
20150814 11:57:36< travis-ci> wesnoth/wesnoth#7037 (master - 0d6fdfe : lipk): The build is still failing.
20150814 11:57:37< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/75588112
20150814 11:57:37-!- travis-ci [~travis-ci@ec2-54-144-26-151.compute-1.amazonaws.com] has left #wesnoth-dev []
20150814 12:01:37< Aginor> wesnoth webserver(s) seem to be down
20150814 12:04:51< lipkab> Aginor: Great! I don't have the necessary administrative powers, however, so you'll have to wait a bit for the technicalities to be done.
20150814 12:04:55< lipkab> shadowm: ^
20150814 12:05:02< Ravana_> seems back
20150814 12:05:10< lipkab> As for the scrolling issue, I'm listening.
20150814 12:08:19< Aginor> Thanks! :)
20150814 12:08:54< Aginor> I was messing around a bit, and I'm still trying to get my head around the entire rendering engine
20150814 12:09:15< Aginor> but it looks like it's actually the first full row of tiles that is repeated during the scroll
20150814 12:10:14< Aginor> forcing a full redraw by using the glibc workaround fixes the symptop (rendering looks fine)
20150814 12:10:30< Aginor> but it doesn't resolve the underlying cause
20150814 12:10:46< Aginor> when you looked into it, do you remember what your theories were?
20150814 12:11:47< lipkab> Well, I recall only one idea.
20150814 12:12:31< lipkab> When scrolling, most of the screen isn't redrawn.
20150814 12:13:28< lipkab> Instead, the engine copies the area that remains on screen to its new location.
20150814 12:14:20< lipkab> My idea was that somehow it tries to copy from or paste to an off-screen area.
20150814 12:14:39< lipkab> It would have explained a number of things:
20150814 12:15:52< lipkab> 1) The behaviour in this case could very well change with library versions.
20150814 12:17:25< lipkab> 2) Scrolling in different directions requires different calculations with regards to which area to copy and where to paste it, so it's not that weird why only direction is messed up.
20150814 12:18:18< lipkab> 3) It could easily be a one-pixel mistake thus nobody noticed it yet.
20150814 12:19:12< lipkab> Unfortunately, I couldn't find how, if in any way, the involved calculations are broken.
20150814 12:27:22-!- Appleman1234 [~Appleman1@KD106179127109.au-net.ne.jp] has joined #wesnoth-dev
20150814 12:29:31< Aginor> lipkab: thank you
20150814 12:30:11< Aginor> I still need to wrap my head around more of that code, but it kind of confirms my suspicions
20150814 12:30:25-!- jcnewjersey [~jcnewjers@pool-108-35-38-39.nwrknj.fios.verizon.net] has joined #wesnoth-dev
20150814 12:31:43< Aginor> it could also be a previously unnoticed overflow in height/width in the rectangle calculations since signedness has been introduced
20150814 12:31:53< Aginor> but that's unlikely
20150814 12:35:00< lipkab> Yeah.
20150814 12:36:33-!- prkc [~prkc@4E5CC75C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20150814 12:38:36< Aginor> anyway, it's getting late for me
20150814 12:38:45< Aginor> I'll look at it more tomorrow
20150814 12:39:00< Aginor> see if I can't instrument some of that rendering code a bit
20150814 12:39:20< lipkab> Ok, good night.
20150814 12:42:21< Aginor> have a good day :)
20150814 12:44:32-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has joined #wesnoth-dev
20150814 12:58:17-!- boucman_2 [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20150814 13:00:43-!- prkc [~prkc@4E5CC75C.dsl.pool.telekom.hu] has quit [Read error: Connection reset by peer]
20150814 13:01:35-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 244 seconds]
20150814 13:02:18-!- sailorswift [~sailorswi@c-24-5-126-97.hsd1.ca.comcast.net] has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
20150814 13:02:33-!- lipkab [~the_new_l@host-91-147-210-193.biatv.hu] has quit [Quit: Leaving]
20150814 13:24:48-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150814 13:24:48-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150814 13:24:48-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150814 13:32:59-!- Kwandulin [~Miranda@p5B0080CB.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150814 13:41:47< wedge009> shadowm: Ctrl+j still works because that's when both character is 'j' and key code is j. (And yes, I know it's the default for Objectives because that's what kept coming up every time I tried to use Ctrl+Return/Enter.) When Ctrl+Return/Enter is used, character is 'j' but key code is Return/Enter, so we should use that key instead.
20150814 13:42:14< wedge009> shadowm: Ctrl+Shift+j still resolves to Ctrl+J.
20150814 13:43:18< wedge009> shadowm: I used the same logic for Ctrl+m, but I haven't tested it on Linux (and I don't have a Mac). But my guess is that the result should be the same as Ctrl+j is for Windows.
20150814 13:48:01< wedge009> Aginor and shadowm: I compiled with SDL2 on Windows a while back. I only needed to make some updates to src/desktop/windows_tray_notification.cpp. I also have a few other minor updates for Windows as well. Shall I make a pull request for those as well?
20150814 13:51:35-!- prkc [~prkc@4E5CC75C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20150814 13:52:18-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Ping timeout: 246 seconds]
20150814 13:52:58-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Ping timeout: 272 seconds]
20150814 13:53:56< wedge009> Oh, I see. Aginor's gone to bed. I'll be going as well. I suppose shadowm is away.
20150814 14:16:42-!- irker035 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150814 14:40:57-!- Xara [~Yangyf@2401:de00:1:6:78f6:845a:6a8a:cee5] has quit [Read error: Connection reset by peer]
20150814 14:48:51-!- Grickit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20150814 15:17:19-!- N4tr0n [~user@cpe-76-175-70-130.socal.res.rr.com] has quit [Remote host closed the connection]
20150814 15:18:46-!- N4tr0n [~user@cpe-76-175-70-130.socal.res.rr.com] has joined #wesnoth-dev
20150814 15:25:38< celticminstrel> wedge009: I'm not entirely sure what you're saying, but in 1.12 on Mac when I press Ctrl-J or Ctrl-M in the hotkey editor I get Ctrl-J and Ctrl-M.
20150814 15:29:21< celticminstrel> Aginor: (void) foo; works in clang and Visual Studio (at least 2015, but probably earlier as well); there are a couple instances of it in BoE.
20150814 15:30:42< iceiceice> celticminstrel: yes, i believe that (void) foo is totally portable
20150814 15:30:55 * celticminstrel nods.
20150814 15:30:56< iceiceice> b/c it actually works in C also iiuc
20150814 15:31:38< celticminstrel> For unused function parameters though I usually omit the name instead.
20150814 15:32:02-!- Kwandulin [~Miranda@p5B0080CB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150814 15:32:23< iceiceice> i remember there was some wierd case in our code where, a variable was unused if a certain #define was present but not otherwise
20150814 15:32:34< iceiceice> and then void foo; is what you want 
20150814 15:33:56< celticminstrel> You forgot the parentheses. :P
20150814 16:03:11-!- Appleman1234 [~Appleman1@KD106179127109.au-net.ne.jp] has quit [Ping timeout: 256 seconds]
20150814 16:04:52-!- Kwandulin [~Miranda@p5B0080CB.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150814 16:09:04< iceiceice> :p
20150814 16:41:56-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has joined #wesnoth-dev
20150814 16:49:01-!- [Relic] [~Relic]@2602:306:33a3:6d30:888c:dec1:f49e:3428] has joined #wesnoth-dev
20150814 16:53:02-!- zombah [~zombah@2a02:28:3:1:214:4fff:fe47:5920] has quit [Quit: Leaving]
20150814 16:53:32< celticminstrel> Which is the better tag-name for AMLAs? [advance] or [advancement]
20150814 16:54:20< vultraz> Latter
20150814 16:55:30< celticminstrel> Does it seem reasonable to support require_amla=!some_id to make two AMLAs mutually exclusive? (I suppose to make them truly exclusive the other would have to also have require_amla=!other_id.) Or would some different syntax (new key?) be better?
20150814 16:56:22< vultraz> !some_id?
20150814 16:56:23< shikadibot> vultraz: Sorry, I don't know what "some_id" means...
20150814 16:56:54< celticminstrel> require_amla=some_id means that the AMLA with id=some_id is a prerequisite for this one.
20150814 16:57:04< celticminstrel> Adding ! negates that.
20150814 16:57:10< vultraz> oh, that's a thing?
20150814 16:57:14< celticminstrel> Or rather, could negate that. It doesn't work that way right now.
20150814 16:57:30< vultraz> So it means
20150814 16:57:37< vultraz> as long as you DON'T have some_id?
20150814 16:57:46< celticminstrel> Yes.
20150814 16:57:51< vultraz> Hm
20150814 16:58:11< vultraz> It would be nice, yes
20150814 16:58:12< iceiceice> celticminstrel: it might be a little less cryptic if you just make a new attribute
20150814 16:58:21< vultraz> but a bit outside the scope of WML
20150814 16:58:31< iceiceice> =! is not valid syntax in any language that i'm aware of
20150814 16:58:33< celticminstrel> What do you mean, vultraz?
20150814 16:58:44< vultraz> we usually use new keys for that kind of stuff
20150814 16:58:48< iceiceice> i guess unless you coundt like bool b = !c
20150814 16:59:06< iceiceice> our coding style usually enforces a space after = thoguh
20150814 16:59:07< celticminstrel> So then maybe exclude_amla=?
20150814 16:59:52< celticminstrel> iceiceice: The ! form would treat ! as an unary operator, so eg require_amla=prereq,!exclude would work too.
20150814 16:59:55-!- Appleman1234 [~Appleman1@KD036009090246.au-net.ne.jp] has joined #wesnoth-dev
20150814 17:00:06< iceiceice> what if its like a comma separated list though
20150814 17:00:31< celticminstrel> Excluding more than one would be require_amla=!one,!two
20150814 17:00:52< iceiceice> so what is the meaning of like "require_amla = one, !two, three, !four"
20150814 17:01:11< iceiceice> does the not apply to each individual entry or all subsequent entries ?
20150814 17:01:19< celticminstrel> Each individual entry.
20150814 17:01:34< iceiceice> yeah so i would agree that's more natural but soe people in this project would assume the other way
20150814 17:01:47< iceiceice> for isntnace in our terrain code stuff the + and - is more like "all subsequent entries"
20150814 17:01:54< celticminstrel> So does exclude_amla sound good?
20150814 17:03:15-!- boucman_2 [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Ping timeout: 256 seconds]
20150814 17:03:57< iceiceice> it sounds good to me anyways
20150814 17:04:04< iceiceice> bbl
20150814 17:04:07-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20150814 17:09:17-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150814 17:19:15-!- ancestral [~ancestral@75-161-229-81.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150814 17:22:46< Ravana_> when doing it with ! then it could use same format as terrain= , thought new key would be better
20150814 17:53:29-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150814 17:54:42-!- {V} [~V@105-70-ftth.on.nl] has quit [Ping timeout: 255 seconds]
20150814 17:57:35-!- {V} [~V@105-70-ftth.on.nl] has joined #wesnoth-dev
20150814 17:59:19-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150814 18:03:28-!- Grickit is now known as Gambit
20150814 18:13:28-!- lipkab [~the_new_l@host-91-147-210-193.biatv.hu] has joined #wesnoth-dev
20150814 18:22:56< celticminstrel> Anyway, I've implemented an exclude_amla, now I just need to test it.
20150814 18:33:23< celticminstrel> ...um, why is the AI controlling side 1? o.O
20150814 18:35:24< celticminstrel> It even has an explicit controller=human.
20150814 18:37:08< celticminstrel> This is in campagin mode.
20150814 18:38:15< celticminstrel> And it doesn't happen in HttT, so it must be something I've done...
20150814 18:38:28< celticminstrel> But it also doesn't happen in 1.12.
20150814 18:39:50< celticminstrel> ^campaign
20150814 18:59:25-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 256 seconds]
20150814 19:03:02< celticminstrel> BTW I can never remember which of && and || has higher precedence...
20150814 19:10:17< celticminstrel> So the controller really is set to AI rather than human. I guess the issue is not where I'd hoped.
20150814 19:11:37< celticminstrel> Anyway, exclude_amla works.
20150814 19:13:33< celticminstrel> Uh. Now I get: Failed to convert String "reserved" to type CONTROLLER
20150814 19:15:05< celticminstrel> I didn't change anything from the last time I started the campaign, though on the first time in this session I edited memory to set the controller to human.
20150814 19:15:26< celticminstrel> Maybe that broke it, I dunno.
20150814 19:18:04< celticminstrel> HttT works though, so there must be something wrong with the scenario too, right?
20150814 19:34:32< celticminstrel> ...oh, whoops, I just pushed to master, which means that pull request now has new commits. :/
20150814 19:38:23< celticminstrel> Note to self: Remember to use branches in future.
20150814 19:48:54-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150814 19:49:55-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150814 19:51:14-!- jcnewjersey [~jcnewjers@pool-108-35-38-39.nwrknj.fios.verizon.net] has quit [Read error: Connection reset by peer]
20150814 19:52:27-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 246 seconds]
20150814 19:52:27-!- wedge010 is now known as wedge009
20150814 20:06:37-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150814 20:14:41-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150814 20:22:31-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150814 20:23:12-!- Shackra [~Jorge@186.177.0.7] has joined #wesnoth-dev
20150814 20:41:13-!- Appleman1234 [~Appleman1@KD036009090246.au-net.ne.jp] has quit [Ping timeout: 246 seconds]
20150814 20:43:08< wedge009> celticminstrel: Yes, this is true. Ctrl+j and Ctrl+m resolve to Ctrl+j and Ctrl+m for all operating systems. The problem occurs when one tries to use Ctrl+Return/Enter. From what I understand of the code, there are two values considered when reading keyboard input: the key code (which key was pressed) and the character it resolves to. The issue arises for Ctrl+Return/Enter because the character value is read as Ctrl+j (LF) or
20150814 20:43:44< celticminstrel> Oh. I'll check that too, then.
20150814 20:43:59< wedge009> celticminstrel: Ctrl+m (CR), depending on the operating system. So for that instance, it should be using the key code, not the character value.
20150814 20:44:44< wedge009> Thanks - we've confirmed Windows and Linux, my assumption was that Mac is the same as Linux (being Unix-based now).
20150814 20:44:55< celticminstrel> I confirmed it just now.
20150814 20:45:13< wedge009> What happens?
20150814 20:45:49< celticminstrel> Pressing Ctrl-Return makes Ctrl-M show up. Pressing Ctrl-Enter makes Ctrl-C show up.
20150814 20:46:16< wedge009> Ctrl-C? Uh-oh, that's a new one.
20150814 20:46:18< celticminstrel> Also interesting is that if I press Cmd-Enter/Return, the Cmd key doesn't register.
20150814 20:46:40< wedge009> That's a separate issue, I think.
20150814 20:46:45< celticminstrel> Ctrl-Tab yields Ctrl-i
20150814 20:46:55< wedge009> Oh dear...
20150814 20:47:02< celticminstrel> ...this one's weird, Ctrl-Esc yields Ctrl-{
20150814 20:47:23< celticminstrel> Ctrl-Backspace and Ctrl-Delete work.
20150814 20:48:14< celticminstrel> Ctrl-Space yields Ctrl-`
20150814 20:48:57< wedge009> Ctrl+` is one that I tried to fix as well, as many Ctrl and Ctrl+Shift combinations result in Ctrl+` and Ctrl+@ being read.
20150814 20:49:20< celticminstrel> I didn't try any Ctrl-Shift combos...
20150814 20:49:52< celticminstrel> Ctrl-Shift-Esc --> Ctrl-[
20150814 20:49:55< wedge009> I suspect all these problems arise from using the character value instead of the key code. But from what I read of the comments, the intention behind that was to account for things like Ctrl+:. On US keyboards this is entered as Ctrl+Shift+;. But the comments say this doesn't make sense for a French keyboard where : is entered without a Shift being necessary.
20150814 20:50:17< celticminstrel> Ctrl-Shift-Tab --> Ctrl-Y
20150814 20:50:27< wedge009> In which case, I wonder if this could be better handled by some sort of locale filter. Something that recognises what keyboard is being used.
20150814 20:50:46< celticminstrel> What if it checked the keycode first, character value second?
20150814 20:51:37< celticminstrel> And never gets to the second check if the keycode is a special key (Esc, Tab, Space, Backspace, Delete, Home, End, PgUp, PgDn, Return, Enter, possibly others).
20150814 20:51:49< celticminstrel> (Oh, F-keys.)
20150814 20:52:18< celticminstrel> Ctrl+F-key doesn't work for me (it registers the ctrl but not the F-key, and "right ctrl" shows up).
20150814 20:52:27< celticminstrel> Shift+F-key works though.
20150814 20:52:43< celticminstrel> Ah, wait, that was with F1, but Ctrl-F12 worked. o.O
20150814 20:52:51< wedge009> I think that the key code would always be used then. There would always be a key code available, but not necessarily a character. I think that's what happens when Ctrl+`/@ is read (at least on Windows), because the character value is null and the ASCII table shows ` and @ (and space) related to that.
20150814 20:53:09< celticminstrel> Only Ctrl-F1 is broken, it seems.
20150814 20:53:13< wedge009> Shift isn't usually a problem, I think.
20150814 20:53:23< wedge009> That's crazy.
20150814 20:53:35< wedge009> Anyway, the point is that key combination detection is borked. o.o
20150814 20:53:35< celticminstrel> Why would the key code always be used then?
20150814 20:53:55< wedge009> If you read key code first, and there's always a key code available, it will get read.
20150814 20:54:02< celticminstrel> You'd filter for special key codes first (modifiers plus the ones I've already listed and maybe a few others like Help or Clear).
20150814 20:54:09< celticminstrel> If it's not any of those, you use the character value.
20150814 20:54:21< celticminstrel> Oh! Arrows too. I forgot about arrows.
20150814 20:54:31< wedge009> Hmm.
20150814 20:54:42< wedge009> Maybe when I've woken up properly I'll have a consider.
20150814 20:54:58< celticminstrel> Ctrl-Clear doesn't work for me but I think that's because some other app is intercepting it before it gets to Wesnoth.
20150814 20:55:01< wedge009> What I wrote works on Windows, but no idea what will happen on Linux or Mac.
20150814 20:55:09< wedge009> What's Clear?
20150814 20:55:22< celticminstrel> It's on the numpad in place of Scroll Lock.
20150814 20:55:32< celticminstrel> (I also have Fn in place of Insert.)
20150814 20:55:43< celticminstrel> (Older Macs had Help there.)
20150814 20:56:03< wedge009> Huh, that's confusing. ):
20150814 20:56:45< celticminstrel> When I press Ctrl-Clear after disabling the shortcut in the other app, Ctrl-Numlock shows up, so close enough I guess.
20150814 20:56:48-!- Shackra [~Jorge@186.177.0.7] has quit [Ping timeout: 264 seconds]
20150814 20:57:19< celticminstrel> Fn is uncapturable with SDL. (You can capture it in Cocoa by listening for modifier changes, but you can't capture it as a keypress, I believe.)
20150814 20:57:43< celticminstrel> I misspoke, clear replaces numlock not scroll lock.
20150814 20:57:44< wedge009> Yes, I think it would be.
20150814 20:58:02< wedge009> Every keyboard is different, too. D:
20150814 20:58:07< celticminstrel> I'm not worried about Fn, in any case.
20150814 20:58:47< celticminstrel> Ctrl-F19 shows up as Ctrl-unknown key.
20150814 20:58:57< wedge009> I can only test for what happens in Windows, at any rate. Haven't worked out a Linux compilation at this point.
20150814 20:59:35-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 245 seconds]
20150814 20:59:37< celticminstrel> It only recognizes up to F15.
20150814 20:59:47< celticminstrel> F16-F19 are unknown.
20150814 21:01:30< celticminstrel> Unrelated suggestion: If the hotkeys for titlescreen elements are simple letters, and the letter exists in the button name, underline that letter.
20150814 21:02:30< wedge009> There's the potential for clashes, I would think.
20150814 21:02:36< wedge009> But a nice idea.
20150814 21:02:39< celticminstrel> For example, assuming English and default shortcuts, you'd see Tutorial, Campaign, Multiplayer, Load, Add-ons, Map Editor, Language, Preferences, Credits, Quit.
20150814 21:02:50< celticminstrel> Clashes?
20150814 21:03:36< celticminstrel> In general it'd be good to indicate what the hotkey is in all situations where it's possible.
20150814 21:03:53< Ravana_> I don't get it, http://wiki.wesnoth.org/MP_Tutorial logs show that it has had those images for 6 years, but they seem spam
20150814 21:04:03< celticminstrel> In my opinion, at least.
20150814 21:04:20< celticminstrel> Wow.
20150814 21:05:37< Ravana_> (player referred to that link when asking what bold name means in lobby)
20150814 21:07:30< wedge009> Multiplayer and Map Editor both start with M, as you mentioned.
20150814 21:08:36< celticminstrel> I'm suggesting doing it based on your hotkey settings, so it doesn't matter that they begin with the same letter 0 they have different hotkeys assigned.
20150814 21:08:46< wedge009> All I see in that link are ads. x.x
20150814 21:09:06< wedge009> Oh, I see.
20150814 21:09:27< wedge009> Can submit it as a feature request. (:
20150814 21:10:29 * celticminstrel shrugs.
20150814 21:22:39-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150814 21:37:41-!- Appleman1234 [~Appleman1@KD106179116043.au-net.ne.jp] has joined #wesnoth-dev
20150814 21:39:47-!- lipkab [~the_new_l@host-91-147-210-193.biatv.hu] has quit [Quit: Leaving]
20150814 21:41:56< shadowm> Ravana_: That's what happens when image hosts turn evil. Perhaps you could get somebody to upload new images?
20150814 21:42:07-!- oldlaptop [~quassel@50-107-125-36.adr02.mskg.mi.frontiernet.net] has quit [Ping timeout: 246 seconds]
20150814 21:42:55-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150814 21:43:02< shadowm> wedge009: What kind of updates for Windows?
20150814 22:11:47< Ravana_> I believe I can check it... I have had several ideas about info that mp players should know
20150814 22:14:56-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150814 22:42:01-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150814 22:59:58-!- jcnewjersey [~jcnewjers@pool-108-35-38-39.nwrknj.fios.verizon.net] has joined #wesnoth-dev
20150814 23:01:10-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150814 23:14:38-!- prkc [~prkc@4E5CC75C.dsl.pool.telekom.hu] has quit [Remote host closed the connection]
20150814 23:15:27< Aginor> wedge009: please make a pull requests for those SDL2 things, I'm quite keen on getting SDL2 up to scratch
20150814 23:24:35< Aginor> shadowm: who's looking after travis?
20150814 23:24:59< shadowm> Are tests still broken?
20150814 23:25:57< shadowm> vultraz: Is this still caused by your changes or did you decide it's something else?
20150814 23:27:39< Aginor> https://travis-ci.org/wesnoth/wesnoth/jobs/75588114 is the latest one and it's a fail
20150814 23:27:49< Aginor> tests are timing out
20150814 23:29:07< shadowm> One of the tests in question is taking really long to run here.
20150814 23:30:10< shadowm> So it's not Travis' fault.
20150814 23:30:37< Aginor> ok
20150814 23:33:03< Aginor> https://travis-ci.org/wesnoth/wesnoth/jobs/75211116 appears to be the culprit, that's where tests started failing
20150814 23:33:08< shadowm> It gets stuck doing this forever sometimes and sometimes it doesn't: http://pastebin.com/egswx2XQ
20150814 23:33:47< shadowm> (Lines 94 and 95 ad infinitum.)
20150814 23:33:54< Aginor> that's bad
20150814 23:36:43< shadowm> Checking if it's really vultraz's commit or just a coincidence stemming from the aforementioned non-deterministic test behavior.
20150814 23:42:40< shadowm> I don't think it's vultraz's commit.
20150814 23:44:37< Aginor> no, it doesn't look very determenistic
20150814 23:45:03< Aginor> s/determenistic/related/
20150814 23:45:11< Aginor> I hate it when I type what I read
20150814 23:45:40< shadowm> No wait, it canbe vultraz's commit.
20150814 23:45:51< shadowm> There is a terrain_mask involved and vultraz broke that in another test.
20150814 23:46:21< shadowm> vultraz: I suggest you revert your commits.
20150814 23:46:40-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20150814 23:47:23< shadowm> In fact, it's been long enough that this is no longer a suggestion.
20150814 23:47:57-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150814 23:48:04< Aginor> shadowm: vultraz == Charles?
20150814 23:48:11< shadowm> Charles Dang, yes.
20150814 23:48:16< Aginor> yeah
20150814 23:49:18< shadowm> vultraz: The fact is that whatever you did broke the previous [terrain_mask] behavior and this is not acceptable.
20150814 23:49:38< Aginor> his commit(s) seems to have introduced regressions
20150814 23:49:48< shadowm> vultraz: Especially since from our conversations in another channel you don't seem to know why or how.
20150814 23:50:42< shadowm> I don't believe these tests should use [terrain_mask] for altering single tiles but that's as good as a dedicated terrain_mask test in my view.
20150814 23:51:32< shadowm> I'm going to look into writing a new test that deals with terrain_mask specifically and following 1.12 behavior later.
20150814 23:56:57< shadowm> (As for why the test results are not deterministic: because they are now subject to unit CTH. Apparently the mask is supposed to force unit CTH to be 100%.)
20150814 23:57:14< shadowm> *force unit terrain defense to be 0%.)
20150814 23:57:34< Aginor> shadowm: would it be acceptable to include https://github.com/tcbrindle/sdl2-cmake-scripts into the wesnoth repo?
20150814 23:57:45< shadowm> (Not entirely sure if that's normally enough to make units have 100% CTH each other.)
20150814 23:57:47< Aginor> I'm a bit dubious myself about the licensing of them
20150814 23:57:58< Aginor> and attribution
20150814 23:59:26< shadowm> Aginor: They'd not be the first CMake scripts we use that aren't PD or GPL: https://github.com/wesnoth/wesnoth/blob/master/cmake/COPYING-CMAKE-SCRIPTS
--- Log closed Sat Aug 15 00:00:31 2015