--- Log opened Sun Sep 18 00:00:28 2016
20160918 00:00:57< celticminstrel> Or maybe not?
20160918 00:01:10< celticminstrel> It seems like it's not available in local MP (which does kinda make sense).
20160918 00:10:25< tad_> If I see [ai][/ai][ai][/ai] can I merge them?
20160918 00:10:30-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20160918 00:11:32< tad_> AH. one has TOD and the other not. So I'm guessing no, I can't merge them
20160918 00:12:28< celticminstrel> That's correct.
20160918 00:13:02< celticminstrel> engine, time_of_day, or turns
20160918 00:13:22< tad_> makes sense.
20160918 00:14:21< celticminstrel> Note though that those three keys do not affect any [aspect], [stage], or [goal] tags, or (these won't exist in THoT since they're new) any other tags containing [facet], [value], or value.
20160918 00:14:33< celticminstrel> Oh, or [filter_own] / [filter_enemy].
20160918 00:15:03< celticminstrel> So if those were the only things in the [ai] tags, they would be mergeable.
20160918 00:21:16-!- travis-ci [~travis-ci@ec2-54-211-72-29.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 00:21:17< travis-ci> wesnoth/wesnoth#11049 (master - 0e87d7f : mattsc): The build has errored.
20160918 00:21:17< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160746582
20160918 00:21:17-!- travis-ci [~travis-ci@ec2-54-211-72-29.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 00:26:14< tad_> Gee I hope I have enough memory for this .. THoT S05 .. 40 turns and I'm hanging back to see what happens .. 3 turns in and there's already a lot units out there ..
20160918 00:40:42-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has joined #wesnoth-dev
20160918 01:10:30-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20160918 01:32:23< tad_> Uh oh.
20160918 01:33:03< tad_> Game is hanging on AI turn somewhere
20160918 01:45:55< celticminstrel> Uh-oh.
20160918 01:46:40< tad_> Yeah. I tried turning on log-debug to try to find out where but then it hangs during Load Game.
20160918 01:46:48< celticminstrel> :|
20160918 01:47:28< tad_> I can reload the savegame, move my two units to attack since they're surrounded, and End Turn. A side or two moves then everything stops.
20160918 01:47:47< tad_> System load does not change so it's no a hard loop. Butthat's all I can say
20160918 01:48:54-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160918 01:48:57< tad_> I was playing runandhide on THoT S06 trying to run out turns and see what happened.
20160918 01:56:13< tad_> Uh oh. --log-info=all should log everything, right?
20160918 01:57:18< vultraz> think so
20160918 01:57:26< vultraz> you don't want to turn that on, though :P
20160918 01:58:14< tad_> Well, I did. And it was screaming along. And I clicked End Turn to see where it hangs. So a few things happened and everything stopped. No messages at all.
20160918 02:01:57< mattsc> tad_: you have a savegame to send me?  To see whether it’s the AI …
20160918 02:02:05< tad_> I get one more message, after I Ctrl-C to kill the program: info ai/manager: [composite_ai] for side 2 : Managed AI will be deleted
20160918 02:02:28< mattsc> Uh …
20160918 02:02:30< tad_> I have a saved game, yes.
20160918 02:06:02< tad_> https://github.com/GregoryLundberg/wesnoth/issues/38 .. savegame attached
20160918 02:10:41< mattsc> tad_: doesn’t hang for me; gets through the turn just fine with no noticeable delay
20160918 02:12:09< mattsc> I do get the “UNKNOWN aspect[version*composite_aspect]” message you mentioned earlier though.
20160918 02:12:11< tad_> v1.13.5+dev (55edd99-Clean)
20160918 02:12:24< tad_> Well, that's something.
20160918 02:12:55< tad_> I'm running on Arch, gcc 6.1 and rebuilt a while ago so it's a fairly current build.
20160918 02:13:06< mattsc> I am one commit further than you are (my commit from a couple hours ago)
20160918 02:13:30< mattsc> on OS X
20160918 02:14:11< celticminstrel> Does GCC 4.8 support annotations?
20160918 02:14:13< mattsc> tad_: does it happen no matter what the outcome of your attacks on that Orcish Warrior are?
20160918 02:14:20< celticminstrel> Sorry, they're called attributes.
20160918 02:14:28< celticminstrel> If not, what version does?
20160918 02:14:33< mattsc> I’ve tried a couple times, and the warrior got killed each time.
20160918 02:14:38< tad_> Let me take a pass and not do a thing and see if it hangs
20160918 02:14:46< mattsc> I also tried by just ending the turn.
20160918 02:15:26< tad_> I just did that is it did NOT hang.
20160918 02:15:55< tad_> Let me restart and load it and I'll tell you which tile I hit with whom
20160918 02:17:18< tad_> Aiglondur (28,5) -> Binak (28,6) w/hammer
20160918 02:17:25< mattsc> Ooo, I just got it to crash on me with this:
20160918 02:17:27< mattsc> “Assertion failed: (row < rows_), function val, file /mats/misc/Wesnoth/wesnoth/src/attack_prediction.cpp, line 493."
20160918 02:18:07< tad_> Angarthing (29,5) --> Grol (28,4) w/bolas
20160918 02:18:17< tad_> End Turn
20160918 02:18:33< celticminstrel> Maybe someone has no attacks? I dunno.
20160918 02:18:54< tad_> A unit appears for a side and it hangs. Might be some off-screen AI on AI attack. Dunno. Frozen
20160918 02:18:57< celticminstrel> Huh, apparently my clang actually supports attributes. I thought it didn't.
20160918 02:20:36< tad_> Adn I mean frozen. No painting. no -log-info=all .. only think I can do is sigkill to kill it
20160918 02:20:49< mattsc> tad_: confirmed, just that I get that assertione failure I quoted.
20160918 02:21:07< tad_> I don't see the assertion.
20160918 02:21:42< mattsc> Last person to work on attack_precition.cpp was jyrkive; the Monte Carlo stuff
20160918 02:21:43< tad_> Maybe a compile option. I thought I was using a debug compile. I do have symbols for gdb
20160918 02:22:27< tad_> Just checked and Monte Carlo is enabled .. I guess it's the default now
20160918 02:23:26< mattsc> tad_: yeah, if I turn MC off, the assert goes away for me.
20160918 02:23:28< mattsc> BBR.
20160918 02:23:29< tad_> And I disabled Monte Carlo, replayed my attacks and End Turn and *NO* hang
20160918 02:23:35< mattsc> *BRB
20160918 02:24:52< mattsc> tad_: phew, that means it wasn’t me  ;)
20160918 02:24:53< tad_> .oO(Sometimes I feel bad because I hit so many problems. But, on the other hand, better me, now, than some Steam player next year.)
20160918 02:25:30< mattsc> tad_: we need those troublemaker folks!
20160918 02:25:35< tad_> mattsc: Hey, I feel the same way. My first reaction was to git diff to be sure I didn't change the scene, yet.
20160918 02:26:16< mattsc> tad_: btw, I did the bolas attack, saved the game, and then the crash still happens after reloading that.  So that makes it even easier to reproduce.
20160918 02:26:26< tad_> Well, good.
20160918 02:27:20< tad_> I do hate non-reproducable errors. But, hey, you also have an example of those messages I couldn't reproduce before. So we're saving a stone.
20160918 02:27:47< mattsc> Yeah, I’ll see what I can find out about that.
20160918 02:28:04< tad_> Probably just need to go into the save game stuff and remove the version.
20160918 02:28:30< tad_> I have an old note about it: only happens on Load. Does NOT happen on Back | Turn 29 from the menu.
20160918 02:28:53< vultraz> mattsc: were you able to gather anything more on the save_index lag?
20160918 02:30:07< mattsc> vultraz: not yet really, but it’s next on my list.  Only thing I can say so far is that when you click on “load game”, the svae_index code is called once; when you click on “campaigns”, it’s called a gazillion times and get slower and slower with each call.
20160918 02:30:10< celticminstrel> Wait a second, we're no longer supporting GCC 4.7, right?
20160918 02:30:31< tad_> Gods, I hope not.
20160918 02:30:35< vultraz> celticminstrel: we are not
20160918 02:30:44< tad_> Most 'stable' Linux are shipping 4.8
20160918 02:30:44< vultraz> mattsc: very interesting
20160918 02:30:55< celticminstrel> That means I can remove this ifdef altogether and not worry about the __attribute__(()) syntax.
20160918 02:31:49< celticminstrel> Arrrrgh I just noticed that the enclosing ifdef checked the wrong macro. o.o
20160918 02:31:50 * tad_ chuckles envisioning a mushroom cloud appearing over Celtic Minstrel's computer when he drops support for 4.7
20160918 02:32:50< celticminstrel> That means that Travis wasn't building the GCC builds with constexpr, noexcept, inheriting constructors (which I think we don't use yet), or ref-qualifiers.
20160918 02:33:15< celticminstrel> It was a type - ifdef __GNUX__ instead of __GNUC__ :(
20160918 02:33:19< celticminstrel> ^typo
20160918 02:34:12< celticminstrel> Okay, now I've changed global.hpp so it'll be almost a full rebuild... :/
20160918 02:35:40< celticminstrel> And there's a linker warning of "main already defined in lua.obj, second definition will be ignored". That seems bad, so I'm forcing a full rebuild now. >_>
20160918 02:35:54< celticminstrel> Geh, it's still there. :|
20160918 02:35:59< celticminstrel> What.
20160918 02:36:02< celticminstrel> Why.
20160918 02:36:04< celticminstrel> :|
20160918 02:36:11< tad_> 'make clean; make all'
20160918 02:36:32< celticminstrel> I'm pretty sure that's what "Rebuild solution" does, but I could try explicitly cleaning and then explicitly building.
20160918 02:36:37< tad_> I try to only do that when I'm headed to bed.
20160918 02:36:47< vultraz> mattsc: wait
20160918 02:37:00< celticminstrel> Apparently F15 is the break key.
20160918 02:37:22< celticminstrel> I think.
20160918 02:37:42< vultraz> mattsc: can you try reverting 9a569df848f3b7d774ac29c100a4f36c4923b7bd and see if it still happens?
20160918 02:37:57< celticminstrel> Nope, cleaning first did not help. The warning's still there, :|
20160918 02:40:18< celticminstrel> So the solution is presumably to exclude src/lua/lua.cpp from the build. Why wasn't that done already?
20160918 02:40:36< celticminstrel> (Admittedly I haven't actually tried it yet, maybe it causes other problems or something.)
20160918 02:41:19 * tad_ is frantically searching YouTube for an appropriate mushroom cloud video, with sound ...
20160918 02:41:26< celticminstrel> Looks like it's not even mentioned in the XCode project, maybe I should remove the reference entirely from the MSVC project?
20160918 02:41:35< vultraz> celticminstrel: i build lua as a separate target
20160918 02:41:40< vultraz> is that not how it is in xcode?
20160918 02:41:42< celticminstrel> So does MSVC.
20160918 02:42:00< celticminstrel> XCode has everything in a giant monolithic target (except wesnothd and unit tests which are separate targets).
20160918 02:42:11< celticminstrel> I want to change that but it's such a lot of work.
20160918 02:42:26< celticminstrel> Though I could easily extract lua into a separate target, at least.
20160918 02:44:49< celticminstrel> Why can't MSVC detect that assert(false || "") means the function doesn't return. :|
20160918 02:44:57< celticminstrel> (Placed at the end of the function.)
20160918 02:45:23< vultraz> why not assert(true)?
20160918 02:45:37< celticminstrel> Huhwhat? I want an unconditional assert here.
20160918 02:45:41< celticminstrel> So that it always asserts.
20160918 02:45:52< celticminstrel> I could always just throw 42 or something, I guess.
20160918 02:45:53< tad_> assert (false && "") ???
20160918 02:45:55< vultraz> how does assert(true) not always assert :|
20160918 02:46:04< vultraz> oh wait
20160918 02:46:05< celticminstrel> assert(true) never asserts.
20160918 02:46:07< vultraz> .. :|
20160918 02:46:11< celticminstrel> tad_: What?
20160918 02:46:25< tad_> false AND a truthy 
20160918 02:46:45< celticminstrel> tad_: It's an idiom I saw somewhere for adding a useful error message to the assert.
20160918 02:46:59< celticminstrel> I just didn't bother typing out the whole error message.
20160918 02:47:02< tad_> false OR true won't work
20160918 02:47:15< tad_> change || to &&
20160918 02:47:32< celticminstrel> tad_: Whoops, typo. I had && in the source but mistyped it over here.
20160918 02:47:41< tad_> oh, good
20160918 02:48:36< celticminstrel> I guess I'll change it to throw "message" for now, whatever.
20160918 02:49:45< tad_> I should add some [if] logic to this scene .. make the AI set village value to 0 after a few times it gets ambushed by a capture. 40 turns it it still kills the peasants then takes the village to heal and spawns more.
20160918 02:51:22< tad_> Which is probably why Monte Carlo died .. too many units fighting too many others.
20160918 02:57:29< mattsc> vultraz: will try
20160918 02:58:47< mattsc> vultraz: can’t revert that (easily); conflicts
20160918 02:59:00< vultraz> mattsc: hm.. hang on
20160918 03:00:06< vultraz> mattsc: ok just try commenting out..game_initialization/create_engine.cpp:1193-1201
20160918 03:01:07 * mattsc is trying that
20160918 03:04:38< mattsc> vultraz: yep, that does it  (clap)
20160918 03:04:56< vultraz> well, that's good 
20160918 03:05:09< vultraz> but it also means that feature I introduced is buggy :/
20160918 03:05:22< mattsc> No problem when I click on the campaigns button any more; save_index does not get created until you click on ‘load game’.
20160918 03:05:29< mattsc> But that’s probably okay, right?
20160918 03:05:55< mattsc> I need to be off for a while and read about … Vikings, I think
20160918 03:06:38< vultraz> I shall have to revert that commit
20160918 03:09:31< celticminstrel> tad_: Out of curiosity, I wonder if using a sparse array in damage calculations would make any difference?
20160918 03:09:45 * tad_ shrugs.
20160918 03:10:22< tad_> Not looked at it. Somehow I wonder what's wrong with Monte Carlo .. seems it should be insensitive to the size of the problem.
20160918 03:11:02< celticminstrel> Firefox, why on earth do you replace the entire contents of the URL bar when I drag something to it that's not a valid URL.
20160918 03:15:27-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160918 03:25:47< mattsc> vultraz: :(
20160918 03:25:58< vultraz> mattsc: hm?
20160918 03:26:36< mattsc> Just commiserating 
20160918 03:27:38< mattsc> tad_: Well, I am getting an actual error leding to an assert, so it might be a bug rather than an overtaxing of the MC.
20160918 03:27:46< mattsc> *leading
20160918 03:29:04< tad_> As I said, I've never looked at that area. I was thinking I might try walking down the number of units on the board. But it seems unnecessary since you're getting a clue there.
20160918 03:29:37< mattsc> tad_: yeah; you could also just try to debug kill half of the units on the field and see what happens.
20160918 03:29:45< tad_> one moment
20160918 03:30:35< mattsc> In any case, does anybody know what jyrkive reads the logs?  This should be reported to him.
20160918 03:30:48< mattsc> s/what/if
20160918 03:30:56< mattsc> Sigh; really can’t type tonight ...
20160918 03:33:09< tad_> That did it. There was a Bandit fighting an Orc to the NW corner. Killed the Bandit with debug. My leader killed the unit to his S (which he does occasionally and it still hangs) and with one less unit on the board in a fight it did NOT hang.
20160918 03:33:37< tad_> Repeating with a unit not in a fight ...
20160918 03:33:58< vultraz> he does read logs
20160918 03:35:19< tad_> Sure seems like something stressing the MC logic. Disable it and it runs. Kill a unit NOT in a fight, hangs. Kill a unit in a fight, runs.
20160918 03:37:19< tad_> I'm amazed I make it through TSG without hitting this. That Long March scene generates a LOT of units. Most of them, though are lost bumping into each other in a mass and never get to fight .. so maybe that's why ..
20160918 03:37:50< celticminstrel> Well, this isn't much of an improvement. Sure it draws, but now it's trapping all input, so I can't even scroll, except it's not really trapping it 'cause I can't type either.
20160918 03:38:38< tad_> Sounds like my computer when I'm linking Wesnoth for a debug build and have Chrome up wasting time waiting for the build to finish.
20160918 03:38:39< celticminstrel> Huh? Somehow when I switched my monitor back it was gone. o.o
20160918 03:39:14< celticminstrel> And I can scroll, but other input is still trapped.
20160918 03:39:30< vultraz> celticminstrel: sounds like the dispatcher is in the wrong place
20160918 03:39:58< celticminstrel> Can't even answer yes in the "do you want to quit" dialog.
20160918 03:40:16< celticminstrel> Or so I thought, but I guess it was just slow?
20160918 03:40:47< celticminstrel> vultraz: Theoretically, as a tpopup, it's not supposed to be doing anything with key or mouse events (though I'll need to change that).
20160918 03:46:50< celticminstrel> Aha, the "allow interaction" flag is false so it's treating it as a tooltip rather than a non-modal.
20160918 03:57:18-!- fabi_ [~fabi@176.5.123.186] has quit [Remote host closed the connection]
20160918 03:57:43-!- fabi_ [~fabi@176.5.123.186] has joined #wesnoth-dev
20160918 04:17:17< vultraz> hm
20160918 04:17:22< vultraz> I wonder if I do need a class for each row
20160918 04:18:59< vultraz> or mayb not
20160918 04:29:19-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160918 04:31:58-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160918 04:41:46 * vultraz mutters darkly
20160918 04:41:56< vultraz> GUI1 COLOR CODES HARDCODED IN A GETTER :|
20160918 04:42:38< celticminstrel> ?
20160918 04:45:00< vultraz> each side has a vector of strings of the available colors, and the gui1 color formatting is baked into that string :|
20160918 04:46:44< shadowm> >says 'mutter'
20160918 04:46:50< shadowm> >proceeds to shout as loudly as possible
20160918 04:51:19-!- Kwandulin [~Miranda@p200300760F2C716E646B7F8E3FD92B51.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 04:52:45< vultraz> celticminstrel: so im curious, what does it mean when you get a 'use of deleted function' error
20160918 04:53:29< shadowm> That you tried to use a function that was marked = delete?
20160918 04:54:00< shadowm> Like trying to copy-construct and object whose copy constructor was deleted to protect it against incompatible semantics.
20160918 04:54:20< shadowm> (e.g. unique_ptr)
20160918 04:54:29< vultraz> I see
20160918 04:57:18< celticminstrel> What he said.
20160918 05:02:09< vultraz> blah i have to do this :|
20160918 05:02:12< vultraz> 			const std::string c = color.substr(color.find_first_of(">") + 1);
20160918 05:02:13< vultraz> 			std::string cid = c;
20160918 05:02:15< vultraz> 			cid[0] = std::tolower(cid[0]);
20160918 05:02:32 * vultraz makes note to look into using an index for RC-ing here
20160918 05:02:36< celticminstrel> I don't really get it?
20160918 05:02:47< celticminstrel> Index? Huh?
20160918 05:03:12< vultraz> celticminstrel: i'm making the color dropdown display a little colored orb
20160918 05:03:37< celticminstrel> So, what's the contents of color?
20160918 05:03:55< vultraz> the vector has a hardcoded "<formatting>Color", so I need to strip the formatting, then make a copy and use tlower on the first char so it works with ~RC
20160918 05:03:57< vultraz> :/
20160918 05:04:03< vultraz> s/has a/is
20160918 05:04:13< celticminstrel> What is this vector and why do you need to use it?
20160918 05:04:24< vultraz> something in connect engine
20160918 05:04:32< vultraz> I'll look to clenaning it up later
20160918 05:04:34< vultraz> cleaning 
20160918 05:04:43< celticminstrel> Why do you need to use this vector?
20160918 05:04:55< vultraz> good question
20160918 05:05:03< Aginor> hey everyone
20160918 05:05:27< celticminstrel> Hi.
20160918 05:05:35< vultraz> ah, Aginor 
20160918 05:05:35< shadowm>    const std::string c = color.substr(color.find_first_of(">") + 1);
20160918 05:05:46< celticminstrel> Do you think it's bad that events might be processed out of order?
20160918 05:05:51< vultraz> Aginor: don't forget to vote :)
20160918 05:05:59< shadowm> Is color.find_first_of(">") guaranteed to not return npos or the index preceding the end of the string?
20160918 05:05:59< Aginor> vultraz: I've just made a post :/
20160918 05:06:01< Aginor> https://forums.wesnoth.org/viewtopic.php?f=2&t=44616&p=602418#p602418
20160918 05:06:01< celticminstrel> Specifically if events::pump() is called during events::pump().
20160918 05:06:34< Aginor> I'm forced to step down even before I'm elected :/
20160918 05:06:44< vultraz> Aginor: ah, I see. Well, you're still eligible to cast a vote for one of the other candidates. 
20160918 05:06:51< shadowm> (i.e. given C = color.find_first_of(">"); C != npos && C < color.length() - 1)
20160918 05:06:51< Aginor> indeed, and I will
20160918 05:06:58< celticminstrel> shadowm: Not generally guaranteed, but if it's a private vector where you know every string contains >
20160918 05:07:10< Aginor> celticminstrel: it's not ideal...
20160918 05:07:12< shadowm> celticminstrel: Iwas  asking vultraz.
20160918 05:07:25< shadowm> Or you know the context then?
20160918 05:07:27< Aginor> celticminstrel: especially not with regards to things like input events
20160918 05:07:29< vultraz> in this case it's guaranteed that it's not npos.
20160918 05:07:39< shadowm> What about the second condition?
20160918 05:08:02< vultraz> also true,
20160918 05:08:05< celticminstrel> ie, do you know that the string doesn't end with >
20160918 05:08:11< Aginor> celticminstrel: for example, reading input from the keyboard, or processing mouse location events, could be rather fraught with difficulties
20160918 05:08:14< celticminstrel> (Unless it has more than one > )
20160918 05:08:43-!- JyrkiVesterinen [~JyrkiVest@87-100-209-188.bb.dnainternet.fi] has joined #wesnoth-dev
20160918 05:08:56< celticminstrel> Aginor: From how it works I get the impression that this is the case now, since it seems that events::pump pops all events from the queue and then processes them in its own time.
20160918 05:09:38< celticminstrel> I can't seem to find the vector vultraz is complaining about. Oh well.
20160918 05:09:57< Aginor> celticminstrel: indeed
20160918 05:10:14< Aginor> celticminstrel: I suspect that we are to some extent lucky
20160918 05:10:20< celticminstrel> How so?
20160918 05:10:46< Aginor> celticminstrel: as it's primarily new GUI components that pump events, and the elements that do so claim full control of the screen as they exist
20160918 05:10:54< celticminstrel> Ah.
20160918 05:11:40< celticminstrel> Do new GUI components need to pump events?
20160918 05:12:06< Aginor> that's a great question
20160918 05:12:21< Aginor> and currently, yes
20160918 05:12:34< Aginor> as we have no main part that's responsible for driving the overall engine
20160918 05:13:04< Aginor> the closest thing is the play_controller classes, but they are not high enough in the hierarchy to be able to do the job of everything
20160918 05:13:14< celticminstrel> I guess normally they wouldn't return to the outer pump until the window is closed.
20160918 05:13:27< Aginor> s/of everthing/for everything/
20160918 05:13:27< Aginor> indeed
20160918 05:13:27< celticminstrel> At which point the events being processed are kind of redundant.
20160918 05:13:49< Aginor> not at all, you want the outer process to resume operation
20160918 05:13:58< Aginor> and event pumping is rather important to that
20160918 05:14:14< celticminstrel> Ah, maybe.
20160918 05:14:32< celticminstrel> twindow::show_non_modal however pumps events and then immediately returns to the outer pump.
20160918 05:14:36< celticminstrel> Maybe it just shouldn't pump events?
20160918 05:15:00< celticminstrel> I suppose I can try that tomorrow.
20160918 05:15:22< celticminstrel> (Also, the event queue seems to get flooded with timer events sometimes.
20160918 05:15:23< celticminstrel> )
20160918 05:16:36< Aginor> that's probably because gui2 is really slow ;)
20160918 05:17:06< Aginor> I have some ideas of things we should do to make event and drawing easier in the engine
20160918 05:17:17< Aginor> but some of them are reasonably major changes
20160918 05:18:05< JyrkiVesterinen> 20160917 20:22:13< gfgtdf> celticminstrel: i actually wonder why you have msvc 2013? i thought that when you installed it the 2015 version was already avaialbe.
20160918 05:18:16< JyrkiVesterinen> I think it's a good thing to have the 2013 version.
20160918 05:18:43< celticminstrel> I'll also note since it was brought up again that I probably installed it in 2014 or something, implying 2015 wasn't yet out.
20160918 05:18:44< JyrkiVesterinen> I have both, and I use 2013 to prevent myself from accidentally using C++11/14 features that 2013 doesn't support.
20160918 05:19:50< Aginor> how is the work getting on for slaying GUI1?
20160918 05:20:04< celticminstrel> Getting very close. I'm currently working on the floating textbox.
20160918 05:20:18< celticminstrel> (Which is why twindow::show_non_modal )
20160918 05:20:22< Aginor> at which point we can remove all of the GUI1 classes?
20160918 05:20:37< celticminstrel> Roughly speaking, once the help dialog is ported.
20160918 05:20:46< celticminstrel> (I say that because most likely it'll be the last one.)
20160918 05:21:03< celticminstrel> Though there's also the in-game theme to account for. That shouldn't be too hard, I think, but...
20160918 05:21:08< Aginor> what about the main gamescreen?
20160918 05:21:15< Aginor> had that one been turned into gui2 yet?
20160918 05:21:18< celticminstrel> Yeah, that's what I meant just there.
20160918 05:21:36< celticminstrel> There are about five dialogs left to port if I recall correctly.
20160918 05:21:48< Aginor> excellent
20160918 05:21:55< celticminstrel> Help, MP Connect, stats, attack prediction, I think I'm missing at least one.
20160918 05:22:11< celticminstrel> And in-game UI / storyscreen needs to become GUI2 somehow.
20160918 05:22:22< celticminstrel> Maybe storyscreen was the one I was missing.
20160918 05:22:28< Aginor> storyscreen is completely separate from game ui
20160918 05:22:38< celticminstrel> I imagine a number of GUI1 widget classes could already be removed.
20160918 05:22:53< celticminstrel> Yeah, I know it is, but it's similar to in-game UI in the use of floating widgets.
20160918 05:22:53< Aginor> it should be pretty easy to port, especially if the text effect is removed
20160918 05:23:11< celticminstrel> I think the main difficulty is that apparently the panel can move from page to page.
20160918 05:23:13< Aginor> just make it a big fat window on top of everything else ;)
20160918 05:23:22< Aginor> the panel?
20160918 05:23:29< celticminstrel> Where the text is.
20160918 05:23:34< celticminstrel> vultraz claims it can move.
20160918 05:23:36< Aginor> hmm
20160918 05:23:37< celticminstrel> I dunno.
20160918 05:23:40< Aginor> I'm not sure it can
20160918 05:23:43< vultraz> what?
20160918 05:23:49< celticminstrel> Storyscreen.
20160918 05:23:50< vultraz> what are we talking about?
20160918 05:23:58< vultraz> yes, the text can be put in different locations
20160918 05:24:11< vultraz> NOT implying you can drag it or something
20160918 05:24:15< celticminstrel> vultraz: Different locations per page, or across the whole story?
20160918 05:24:21< vultraz> per page
20160918 05:24:41< celticminstrel> I think all you need to do to enable that is use set_origin() and place(), but there may be more to it.
20160918 05:25:02< celticminstrel> Maybe just set_origin() would be sufficient.
20160918 05:25:19< celticminstrel> (These are widget methods.)
20160918 05:26:07< celticminstrel> When I tried the floating textbox as an actual floating GUI2 widget, it did seem like set_origin() and place() correctly set its location, but since other stuff wasn't working I'm still not clear if that's sufficient overall.
20160918 05:26:55< celticminstrel> I did consider that perhaps just the panel could be made into a GUI2 dialog, which is then displayed above the map+journey+title.
20160918 05:27:23< Aginor> no, it'd be much cleaner to make it fit within the GUI framework
20160918 05:27:52< vultraz> yeah
20160918 05:27:55< vultraz> I'll make it an actual dialog
20160918 05:28:03< vultraz> but I'm postponing it for now
20160918 05:28:04< celticminstrel> Probably clearner.
20160918 05:28:04< Aginor> don't make it a dialog
20160918 05:28:08< celticminstrel> ^cleaner
20160918 05:28:09< Aginor> make it a window
20160918 05:28:14< vultraz> a.. what?
20160918 05:28:19< Aginor> make the main menu a window too
20160918 05:28:25< vultraz> a window?
20160918 05:28:28< vultraz> what is this of which you speak
20160918 05:28:41< Aginor> and the main gamescreen
20160918 05:28:41< celticminstrel> But if you're making the whole thing a GUI2 window, you need to figure out how to handle journeys.
20160918 05:28:43< vultraz> It sounds potentially useful....
20160918 05:28:55< vultraz> I mean, a dialog IS a window
20160918 05:28:58< vultraz> how is this different?
20160918 05:29:00< celticminstrel> Aginor: A dialog is basically a sort of wrapper for a window.
20160918 05:29:21< Aginor> sounds like the terminology is not quite what I expect then :)
20160918 05:29:28< celticminstrel> Quite possibly.
20160918 05:29:39< vultraz> ah
20160918 05:29:43< celticminstrel> For example, twindow extends twidget.
20160918 05:29:47< vultraz> yes I will make it a window
20160918 05:29:52< celticminstrel> (Not directly, I think.)
20160918 05:30:01< celticminstrel> Main menu is already a dialog / window.
20160918 05:30:20< vultraz> but I'm postponing the storyscreen, attack prediction, and stats 
20160918 05:30:26< vultraz> until after 1.13.6
20160918 05:30:29< Aginor> and make the editor a separate window as well
20160918 05:30:30< Aginor> etc
20160918 05:30:39< vultraz> Aginor: eventually, yes
20160918 05:30:53< celticminstrel> Well, the editor is in a sense the same "window" as the main game screen (at the moment), just with a special theme.
20160918 05:30:54< Aginor> that way we can choose to show the aoproriate window based on the outcomes
20160918 05:31:15< vultraz> Aginor: we just need a way to make the gamemap function as a widget and accept the appropriate input
20160918 05:31:18< Aginor> celticminstrel: yes and no...
20160918 05:31:22< vultraz> that's the big problem right now
20160918 05:31:38< celticminstrel> Do we really need to make the game map function as a widget?
20160918 05:31:50< Aginor> it'd be the appropriate thing to do
20160918 05:31:52< vultraz> yes
20160918 05:31:54< celticminstrel> Maybe.
20160918 05:32:02< vultraz> but this is all for later
20160918 05:32:11< vultraz> right now we need to focus on our current projects
20160918 05:32:12< celticminstrel> But...
20160918 05:32:18< Aginor> it would be better than having it as a loose component
20160918 05:32:28< vultraz> it looks like we need to move up deployment of the new addons manager
20160918 05:33:00< celticminstrel> Well, I feel like making it a separate widget implies some sort of major rehaul of like five or six game classes, such as the play controller, the game map class, the various command executors, etc.
20160918 05:33:20< Aginor> yes
20160918 05:33:45< celticminstrel> While it does seem like the "right" way to do it, I kind of question whether it can be ready for 1.14.
20160918 05:34:03< vultraz> we'll see
20160918 05:34:14< vultraz> let's focus on our current goals right now, shall we? :P
20160918 05:34:30< vultraz> if we start looking at that too we'll never get anywhere
20160918 05:38:54-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20160918 05:43:16-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160918 05:43:16-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Read error: Connection reset by peer]
20160918 05:45:40-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160918 05:53:58< JyrkiVesterinen> 20160918 02:17:27< mattsc> “Assertion failed: (row < rows_), function val, file /mats/misc/Wesnoth/wesnoth/src/attack_prediction.cpp, line 493."
20160918 05:54:21< JyrkiVesterinen> Hmm. It means that the HP of a combatant during combat simulation exceeds the maximum HP.
20160918 05:55:09< JyrkiVesterinen> Given that the assertion doesn't happen if Monte Carlo simulation is disabled, the bug is definitely my fault.
20160918 05:55:20< JyrkiVesterinen> I'll have a look next week.
20160918 05:55:56< JyrkiVesterinen> 20160918 02:24:53< tad_> .oO(Sometimes I feel bad because I hit so many problems. But, on the other hand, better me, now, than some Steam player next year.)
20160918 05:56:16< JyrkiVesterinen> Yes, it's definitely better to find the bugs now.
20160918 05:57:14< JyrkiVesterinen> I did test Monte Carlo mode, both with the built-in combat generator in attack_prediction.cpp, and by allowing the AI to play a turn in a build that uses MC simulation for *all* combat simulation.
20160918 05:57:28< JyrkiVesterinen> tad_ must have hit some nasty corner case.
20160918 06:08:53-!- fabi_ [~fabi@176.5.123.186] has quit [Remote host closed the connection]
20160918 06:09:26-!- fabi_ [~fabi@176.5.123.186] has joined #wesnoth-dev
20160918 06:11:07-!- Bonobo [~Bonobo@2001:44b8:254:3200:605a:61d9:df57:9fbd] has quit [Ping timeout: 255 seconds]
20160918 06:11:31-!- Bonobo [~Bonobo@2001:44b8:254:3200:8c4b:c653:8578:9a31] has joined #wesnoth-dev
20160918 06:22:49-!- Bonobo [~Bonobo@2001:44b8:254:3200:8c4b:c653:8578:9a31] has quit [Ping timeout: 255 seconds]
20160918 06:25:28< vultraz> ok, I almost have a working mp staging dialog
20160918 06:25:34< vultraz> well
20160918 06:25:37< vultraz> basically working
20160918 06:25:50< vultraz> haven't gotten chat or the userlist working
20160918 06:28:42-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-dev
20160918 06:30:47-!- Bonobo [~Bonobo@2001:44b8:254:3200:85e4:c67f:8a65:6bf7] has joined #wesnoth-dev
20160918 06:33:20-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160918 06:33:51< JyrkiVesterinen> 20160918 03:35:19< tad_> Sure seems like something stressing the MC logic. Disable it and it runs. Kill a unit NOT in a fight, hangs. Kill a unit in a fight, runs.
20160918 06:34:09< JyrkiVesterinen> Stressing the Monte Carlo logic should not be even possible.
20160918 06:34:24< JyrkiVesterinen> Monte carlo simulation runs in almost constant time. That's the whole point of having it.
20160918 06:35:09< JyrkiVesterinen> The assertion warns about an imminent out-of-bounds array write that can cause all kinds of damage.
20160918 06:35:49< JyrkiVesterinen> In particular, if tad_ is extremely unlucky, the out-of-bounds write may overwrite the iteration counter. That would explain the hang.
20160918 06:36:05< JyrkiVesterinen> (But it's a very unlikely possibility.)
20160918 06:37:26< JyrkiVesterinen> A more likely reason why killing random units stops the game from hanging is that the AI no longer simulates the problematic fight, or simulates it in a different situation (different starting HP or something).
20160918 06:38:50< JyrkiVesterinen> Chaos theory is at play here. A change that appears very insignificant at first glance triggers more and more additional changes... and eventually the AI ends up not simulating a fight at all.
20160918 06:41:09-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Ping timeout: 276 seconds]
20160918 06:44:06-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20160918 06:52:58-!- Bonobo [~Bonobo@2001:44b8:254:3200:85e4:c67f:8a65:6bf7] has quit [Ping timeout: 255 seconds]
20160918 06:53:19-!- Bonobo [~Bonobo@2001:44b8:254:3200:69d9:b8e6:3128:e2b2] has joined #wesnoth-dev
20160918 06:54:59-!- Kwandulin [~Miranda@p200300760F2C716E646B7F8E3FD92B51.dip0.t-ipconnect.de] has quit [Ping timeout: 272 seconds]
20160918 07:18:50< shadowm> loonycyborg: It seems campaignd-1.13 has quit with SIGPIPE before.
20160918 07:19:07< shadowm> (Reality check: the process' exit signal is the exit status AND 127, right?
20160918 07:19:09< shadowm> )
20160918 07:19:43-!- Kwandulin [~Miranda@p200300760F2C71FC646B7F8E3FD92B51.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 07:21:55< shadowm> loonycyborg: Might want to check what happens when something bad happens to either side of the connection while sending an add-on to a user (i.e. client is downloading an add-on).
20160918 07:22:41< shadowm> That's the last action recorded in the log in every case where campaignd appears to have died to SIGPIPE.
20160918 07:43:24-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has joined #wesnoth-dev
20160918 07:43:25-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has quit [Changing host]
20160918 07:43:25-!- iceiceice [~chris@unaffiliated/iceiceice] has joined #wesnoth-dev
20160918 07:46:02-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160918 07:49:33-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 240 seconds]
20160918 07:54:10-!- Kwandulin [~Miranda@p200300760F2C71FC646B7F8E3FD92B51.dip0.t-ipconnect.de] has quit [Ping timeout: 255 seconds]
20160918 08:02:09-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20160918 08:02:09-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20160918 08:02:09-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160918 08:21:20-!- VultCave [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20160918 08:23:14-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Disconnected by services]
20160918 08:23:27-!- VultCave is now known as vultraz
20160918 08:23:31-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20160918 08:23:31-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160918 08:34:45-!- Kwandulin [~Miranda@p200300760F2C71FCF983340CE2D64A97.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 08:44:32-!- JyrkiVesterinen [~JyrkiVest@87-100-209-188.bb.dnainternet.fi] has quit [Quit: .]
20160918 08:45:23< loonycyborg> shadowm: downloading addon is exactly when sendfile is used. I'll read up on it more
20160918 08:46:03-!- iceiceice [~chris@unaffiliated/iceiceice] has quit [Ping timeout: 240 seconds]
20160918 08:47:12< loonycyborg> hmm or it's not particular to sendfile and happens there only because it takes a lot of time..
20160918 08:58:46-!- Kwandulin [~Miranda@p200300760F2C71FCF983340CE2D64A97.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160918 09:04:21-!- Kwandulin [~Miranda@p200300760F2C71FCF99E48F4C9E56368.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 09:05:36-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has joined #wesnoth-dev
20160918 09:14:51< loonycyborg> http://stackoverflow.com/questions/108183/how-to-prevent-sigpipes-or-handle-them-properly
20160918 09:22:16< shadowm> When you do this I always wonder if it means you're trying to teach me something I don't need or...
20160918 09:25:24< shadowm> FWIW according to SDLnet.c, SDL_net just sets SIGPIPE to be ignored.
20160918 09:25:34< shadowm> (Yaaaaaaaay shared libraries tampering with signals.)
20160918 09:26:06< Aginor> yeowch
20160918 09:26:09< Aginor> that's nasty
20160918 09:26:47< shadowm> Like all other SDL underscore-foo libraries, it is only really supposed to be a demo for you to get an idea of how to best use SDL in your application, not a real library.
20160918 09:27:01< shadowm> Didn't stop people from using them as libraries, though.
20160918 09:27:22< shadowm> Plus SDL_net doesn't use SDL.
20160918 09:27:48< shadowm> It uses a couple of utility macros and platform-dependent data type aliases and that's it.
20160918 09:28:50-!- midzer_ [~quassel@p57B45688.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 09:28:52-!- JyrkiVesterinen [~JyrkiVest@78-27-116-159.bb.dnainternet.fi] has joined #wesnoth-dev
20160918 09:29:15< irker094> wesnoth: gfgtdf wesnoth:master f2635e5ad65d / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/f2635e5ad65d991b8911fac1cfc60f88836c3f49
20160918 09:29:56< shadowm> I wonder if there were more pseudo-SDL_foo libraries like that back in the day.
20160918 09:30:05< JyrkiVesterinen> 20160918 09:26:47< shadowm> Like all other SDL underscore-foo libraries, it is only really supposed to be a demo for you to get an idea of how to best use SDL in your application, not a real library.
20160918 09:30:14< shadowm> Like a SDL_string or a SDL_linkedlist.
20160918 09:30:21< JyrkiVesterinen> Yeah. I noticed about a month ago that SDL_mixer supports only one music channel.
20160918 09:30:30< JyrkiVesterinen> We are using it for all audio playback.
20160918 09:30:44< JyrkiVesterinen> We should consider switching to OpenAL or something...
20160918 09:30:49-!- midzer [~quassel@p5B2960C2.dip0.t-ipconnect.de] has quit [Ping timeout: 265 seconds]
20160918 09:30:52< shadowm> That's not a problem for us because we only use one music channel.
20160918 09:31:18< JyrkiVesterinen> Well, crossfading would be nice. It's very hard to support without at least two channels.
20160918 09:31:36< irker094> wesnoth: Gregory A Lundberg wesnoth:master fc4c75e56d65 / data/campaigns/Heir_To_The_Throne/scenarios/23_Test_of_the_Clans.cfg: HttT S23 Fix bug: Unclear objective https://github.com/wesnoth/wesnoth/commit/fc4c75e56d6558689e56a38560d36c9342a74bb2
20160918 09:31:38< irker094> wesnoth: Lari Nieminen wesnoth:master 8397201f60a7 / data/campaigns/Heir_To_The_Throne/scenarios/23_Test_of_the_Clans.cfg: Merge pull request #789 from GregoryLundberg/GL_HttT_S23_objectives https://github.com/wesnoth/wesnoth/commit/8397201f60a7ceb2cce2160b2e306715bea8ff78
20160918 09:31:42-!- midzer_ is now known as midzer
20160918 09:31:58< shadowm> But yes, I've suggested before that using OpenAL or a canned real mixer library would be a vast improvement for some things if done right, and enable us to do more awesome things.
20160918 09:33:01< shadowm> Procedural effects like reverb etc.
20160918 09:33:23< shadowm> Custom looping points for music tracks.
20160918 09:33:53< shadowm> Seeking to arbitrary points.
20160918 09:34:24< shadowm> Not seeing ALSA buffer overrun warnings half of the time.
20160918 09:37:39< shadowm> Oh, and eliminating the random millisecond garbage playback bug when switching tracks on Linux.
20160918 09:38:11< shadowm> Or that issue where Wesnoth just plain crashes for some people on Windows when playing sounds.
20160918 09:50:29< Aginor> that all sounds like rather important goals, especially fixing the random crash
20160918 09:59:24-!- JyrkiVesterinen [~JyrkiVest@78-27-116-159.bb.dnainternet.fi] has quit [Quit: Going somewhere. I'll return in a couple of hours.]
20160918 10:05:04-!- travis-ci [~travis-ci@ec2-107-22-13-18.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 10:05:05< travis-ci> wesnoth/wesnoth#11052 (master - f2635e5 : gfgtdf): The build has errored.
20160918 10:05:06< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160796951
20160918 10:05:06-!- travis-ci [~travis-ci@ec2-107-22-13-18.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 10:10:04-!- Bonobo [~Bonobo@2001:44b8:254:3200:69d9:b8e6:3128:e2b2] has quit [Ping timeout: 255 seconds]
20160918 10:14:12-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160918 10:16:33-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20160918 10:16:34-!- wedge010 is now known as wedge009
20160918 10:23:41< irker094> wesnoth: doofus-01 wesnoth:master 376d45a292fd / / (48 files in 4 dirs): new terrain: Merfolk or aquatic camp (#773) https://github.com/wesnoth/wesnoth/commit/376d45a292fdf2bd24802bb5d6de636189d481f6
20160918 10:33:53-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160918 10:35:48-!- travis-ci [~travis-ci@ec2-54-167-137-80.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 10:35:49< travis-ci> wesnoth/wesnoth#11053 (master - 8397201 : Lari Nieminen): The build passed.
20160918 10:35:49< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160797128
20160918 10:35:49-!- travis-ci [~travis-ci@ec2-54-167-137-80.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 10:50:10-!- Duthlet [~Duthlet@dslb-146-060-179-135.146.060.pools.vodafone-ip.de] has joined #wesnoth-dev
20160918 11:02:35-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 11:02:36< travis-ci> wesnoth/wesnoth#11054 (master - 376d45a : doofus-01): The build passed.
20160918 11:02:36< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160803165
20160918 11:02:36-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 11:12:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160918 11:13:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160918 11:38:08-!- gfgtdf [~chatzilla@x4e363f91.dyn.telefonica.de] has joined #wesnoth-dev
20160918 11:44:52-!- Bonobo [~Bonobo@2001:44b8:254:3200:bc64:9a79:7dc5:557c] has joined #wesnoth-dev
20160918 11:46:58-!- JyrkiVesterinen [~JyrkiVest@78-27-104-98.bb.dnainternet.fi] has joined #wesnoth-dev
20160918 12:04:21-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160918 12:19:03< vultraz> gfgtdf: I have an initial implementation of the connect screen
20160918 12:19:21< gfgtdf> vultraz: so it works already ? :O
20160918 12:19:55< vultraz> almost
20160918 12:20:07< gfgtdf> vultraz: nice
20160918 12:20:16< gfgtdf> vultraz: both mpo connect and mp wait ?
20160918 12:20:47< vultraz> just connect so far but wait should be easy to make use this
20160918 12:33:06-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has joined #wesnoth-dev
20160918 13:02:05-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Remote host closed the connection]
20160918 13:03:30< bumbadadabum> zookeeper: Should we add the merfolk castle to mainline campaigns?
20160918 13:05:56< zookeeper> bumbadadabum, yes, but i haven't really charted suitable places yet. also the actual castle will go in sooner or later anyway, so that should be taken into account too.
20160918 13:06:16< bumbadadabum> DW?
20160918 13:08:07< zookeeper> yes, both will undoubtedly be used a lot in DW
20160918 13:13:23< bumbadadabum> zookeeper: I'm also asking because I wanted to make a FB post showing them off, preferably in DW
20160918 13:13:38< bumbadadabum> and I don't want to promise merfolk castles somewhere that won't have merfolk castles :p
20160918 13:14:21< zookeeper> well maybe wait until the castles get in too, so you can show off both? he'll probably get to finishing those soon enough i'd think
20160918 13:22:30-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160918 13:29:43-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160918 13:31:38< Bonobo> having a non-ruins water castle would be nice for a few multiplayer maps
20160918 13:46:55-!- Kwandulin [~Miranda@p200300760F2C71FCF99E48F4C9E56368.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160918 13:57:43-!- Bonobo [~Bonobo@2001:44b8:254:3200:bc64:9a79:7dc5:557c] has quit [Quit: Leaving]
20160918 14:19:41< vultraz> hm
20160918 14:19:46< vultraz> I wonder if reusing the FLG dialog isn't worth it
20160918 14:19:54< vultraz> and i should just use combos like before
20160918 14:21:34-!- edgrey [~edgrey@178.204.202.173] has joined #wesnoth-dev
20160918 14:22:43< vultraz> it is a rather lot of extra clicks
20160918 14:24:17-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160918 14:30:15-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160918 14:30:25< vultraz> one click for the dialog
20160918 14:30:27< vultraz> one click for faction
20160918 14:30:30< vultraz> 2 for leader
20160918 14:30:38< vultraz> and another for gender..
20160918 14:30:51< vultraz> yeah, this is rather a lot of clicks..
20160918 14:31:28< vultraz> i could use a grid listbox to remove a click for Leader, but the matrix placement is still broken :/
20160918 14:32:20< irker094> wesnoth: mattsc wesnoth:fix-tunnel-issues cc7ae4d65534 / src/pathfind/ (pathfind.cpp teleport.cpp teleport.hpp): Tunnels: fix vision not being propagated through tunnels https://github.com/wesnoth/wesnoth/commit/cc7ae4d65534bd18c327837d633be3de712f9b8e
20160918 14:32:36< vultraz> maybe i could do a box for faction/leader
20160918 14:32:45< vultraz> then 3 toggle buttons for gender..
20160918 14:32:46< mattsc> This ^ is the beginning of PR 790: https://github.com/wesnoth/wesnoth/pull/790
20160918 14:32:48< vultraz> but in which case
20160918 14:32:52< vultraz> why even need the flg dialog :/
20160918 14:33:41-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160918 14:35:24< mattsc> zookeeper, celticminstrel, DeFender1031, gfgtdf, vultraz, all: that first commit in PR 790 so far always transmits vision through tunnels.  Disabling that with a new key to the [tunnel] tag will be next, but I wanted to separate those into their own commits.
20160918 14:37:39< mattsc> Umm, I have no idea how that change in l.127 in teleport.cpp got there.  AFAIR, I didn’t do that …
20160918 14:39:36< celticminstrel> find_routes seems like the sort of function that could really benefit from a parameter pack...
20160918 14:40:03< mattsc> Oh, it’s from gfgtdf’s wml_getter commits.  I guess I did not rebase the branch first ...
20160918 14:40:26< celticminstrel> Line 150 is also familiar.
20160918 14:40:35< celticminstrel> Didn't you already commit that?
20160918 14:40:48< mattsc> yes, I did
20160918 14:41:11< mattsc> I reused the branch from the prvious PR and did not rebase.
20160918 14:41:21< mattsc> My bad.
20160918 14:41:26< celticminstrel> Misspelled vision in the PR title too. >_>
20160918 14:41:30< mattsc> I’ll fix that later.
20160918 14:42:00< mattsc> Grrr.  :P
20160918 14:43:58< mattsc> So, question: since this is a branch off master (I don’t have my own fork), can I force push to that without (risk of)  screwing up the master branch itself?
20160918 14:44:03-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has quit [Ping timeout: 240 seconds]
20160918 14:45:04< mattsc> Or is it better to close this PR when all is done and merge manually (or open another PR)?
20160918 14:45:07< celticminstrel> Yes. In fact, you can force push to master without risk, since it would be rejected. :P
20160918 14:45:21< celticminstrel> No, don't close and re-open unless you absolutely have to.
20160918 14:46:40< mattsc> Okay; I thought the former was the case, but wasn’t entirely sure.
20160918 14:47:04< mattsc> And acknowledged on the latter.  I should learn how to do that anyway…
20160918 14:52:35-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 14:52:36< travis-ci> wesnoth/wesnoth#11055 (fix-tunnel-issues - cc7ae4d : mattsc): The build is still failing.
20160918 14:52:36< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160835182
20160918 14:52:36-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 14:53:01-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Remote host closed the connection]
20160918 14:55:40-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20160918 14:55:55-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160918 14:56:16-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Ping timeout: 255 seconds]
20160918 14:56:23-!- Kwandulin [~Miranda@p200300760F2C71FCAC880FA8F54322DE.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 14:57:27-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160918 15:05:59-!- celmin [~celticmin@bas1-kitchener09-1176181722.dsl.bell.ca] has joined #wesnoth-dev
20160918 15:09:49-!- celmin [~celticmin@bas1-kitchener09-1176181722.dsl.bell.ca] has quit [Changing host]
20160918 15:09:49-!- celmin [~celticmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160918 15:13:43< DeFender1031> mattsc, the previous default is not to have vision through tunnels though. I assume it's your intention to make the default behavior match the existing once you add the flag to turn it off?
20160918 15:13:45-!- prkc [~prkc@gateway/vpn/privateinternetaccess/prkc] has quit [Ping timeout: 248 seconds]
20160918 15:26:41-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160918 15:26:45-!- prkc [~prkc@46.166.190.184] has joined #wesnoth-dev
20160918 15:28:27< mattsc> DeFender1031: no, we’re going to change the default.
20160918 15:28:45< mattsc> Same as for movement.
20160918 15:29:16< mattsc> We discussed that it makes more sense to have movement and vision through tunnels work consistently with all other types of movement and vision, even if it is different from previous behavior.
20160918 15:30:19-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160918 15:32:46< DeFender1031> mattsc, it wasn't clear to me that the conclusion was "change the existing default". I understood it as "this SHOULD HAVE been the default, but for backwards compatibility reasons, it's probably safer to not change it at this point".
20160918 15:34:00< mattsc> DeFender1031: there was another discussion a day or two later in which we changed our mind on that.  I think you weren’t online at the time.
20160918 15:36:12< DeFender1031> mattsc, ah. Seems I ought to start catching up on the backlogs from here too, but there's just so much more to go through here than in the main channel.
20160918 15:36:41< mattsc> DeFender1031: yeah, I know.  I often don’t read everything either, it just takes too long.
20160918 15:47:28 * zookeeper thinks it's possible to frankenstein an adequate silly portrait for bugg the sea orc: https://dl.dropboxusercontent.com/u/63964618/wesnoth/bugg.png (obviously very crude since it's just a test draft)
20160918 15:51:22< vultraz> yeah, toggle buttons don't work
20160918 15:51:24< vultraz> sugh
20160918 15:51:34< vultraz> i don't want to copy the old design verbatim, here ;|
20160918 15:51:52< vultraz> might have to
20160918 15:52:03< vultraz> using the flg manager is just.. too many clicks
20160918 15:53:00< celmin> That's probably mainly the fault of people like me and vultraz who monologue all the time.
20160918 15:56:16< vultraz> celmin: https://drive.google.com/file/d/0B-mR9s8FduLLZVR3Y2lwSURocWM/view?usp=sharing
20160918 15:56:27< vultraz> im thinking ill have to end up using the old design :/
20160918 15:57:10< vultraz> it's just... not working
20160918 15:58:33< vultraz> though i suppose that there .. could.. work
20160918 15:58:48< vultraz> (inb4 but the small resolutions!)
20160918 15:59:18< DeFender1031> zookeeper, I honestly never understood the purpose of bugg the sea orc.
20160918 15:59:37< vultraz> celmin: really need suggestions here
20160918 16:01:39< zookeeper> DeFender1031, comic relief, nothing more
20160918 16:04:23-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160918 16:06:20-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160918 16:06:27-!- nore [~ncourant@sas.eleves.ens.fr] has quit [Ping timeout: 248 seconds]
20160918 16:15:43< celmin> vultraz: You decided not to reuse the flg dialog?
20160918 16:16:20-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has joined #wesnoth-dev
20160918 16:16:36< celmin> Radiobuttons for gender doesn't seem that great, mainly because there's three options so you have that blank cell.
20160918 16:17:11< celmin> Maybe it'd work if you have the three buttons all in one row and use the bottom row for something else?
20160918 16:17:22< celmin> Not sure if there's anything else to put there, though.
20160918 16:17:45< celmin> BTW, you misspelled Wesnoth in the topic.
20160918 16:18:08< celmin> Toggle buttons aside, the layout you just linked looks great.
20160918 16:18:19< vultraz> celmin: the flg dialog is just.. too many clicks
20160918 16:18:29< vultraz> it's 2 clicks to change faction, yes
20160918 16:18:36< celmin> Why is it too many clicks?
20160918 16:18:45-!- Appleman1234 [~Appleman1@KD119104050103.au-net.ne.jp] has quit [Ping timeout: 265 seconds]
20160918 16:18:48< vultraz> but then if you want to change leader, you have to choose your faction again...
20160918 16:18:52< vultraz> then click the dropdown
20160918 16:19:02< vultraz> select a leader..
20160918 16:19:09< vultraz> i suppose i could make the dialog save selections
20160918 16:19:11< celmin> No, don't make them reselect their faction. It should open with all their previous selections already selected.
20160918 16:19:31< celmin> I thought that would be obvious.
20160918 16:20:30< vultraz> im undecided, really, since eliminating extraneous dialogs is always good. with this design, mp wait could get rid of it too and just give players access to the dropdown lists directly..
20160918 16:20:40< vultraz> but using the dialog reduces clutter
20160918 16:21:03< vultraz> if I could eliminate the leader dropdown it would be better
20160918 16:21:13< celmin> I don't really care either way, to be honest.
20160918 16:21:21< vultraz> i wanted to try a 'grid' of leader selections using the grid listbox but matrix palcement is still too broken
20160918 16:21:23< vultraz> placement
20160918 16:21:35< loonycyborg> shadowm: what I really need is some way to reproduce this sigpipe issue in controlled fashion
20160918 16:21:47< loonycyborg> so I could try different fixes..
20160918 16:22:13< celmin> I think I have a vague idea of the problem in matrix placement.
20160918 16:22:23< celmin> get_best_size() doesn't account for linked groups.
20160918 16:22:30< vultraz> likely
20160918 16:22:30< celmin> So it doesn't allocate enough initial space.
20160918 16:22:36< celmin> Thus it gets cut off.
20160918 16:22:53< vultraz> linked groups are rather finicky, I think
20160918 16:23:16< vultraz> sometimes you need to have multiple widgets in a list in a linked group to maintain size
20160918 16:25:31-!- Rhonda [~rhonda@anguilla.debian.or.at] has quit [Changing host]
20160918 16:25:31-!- Rhonda [~rhonda@wesnoth/developer/rhonda] has joined #wesnoth-dev
20160918 16:26:39< vultraz> celmin: also clear is broken with grid listboxes
20160918 16:27:04< vultraz> invalidate layout needs to be called or space is continuously added even if the contents disappear
20160918 16:27:34< vultraz> and there's no way to set row length
20160918 16:29:42< celmin> What's broken?
20160918 16:29:48< celmin> "clear"?
20160918 16:32:24< vultraz> yes
20160918 16:32:26< vultraz> clear()
20160918 16:33:04-!- Kwandulin [~Miranda@p200300760F2C71FCAC880FA8F54322DE.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160918 16:33:30-!- nore [~ncourant@sas.eleves.ens.fr] has joined #wesnoth-dev
20160918 16:35:31< celmin> Setting row length sounds useful...
20160918 16:36:37< vultraz> so Im wondering
20160918 16:37:01< vultraz> could not the grid listbox be implemented using rows of n columns.
20160918 16:37:54< vultraz> though im not sure what the exact implementation would be 
20160918 16:38:39< vultraz> maybe adding 1 generator row
20160918 16:38:43< vultraz> divided into n columns
20160918 16:38:55< vultraz> populated with the defined child
20160918 16:39:06< vultraz> and using spacers for overflow
20160918 16:39:25< vultraz> i think it might be worth making this its own widget
20160918 16:39:36< vultraz> not sure
20160918 16:40:35< celmin> It's a bit hard to specify parameters for the matrix generator that don't exist in the other generators.
20160918 16:44:15-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160918 16:45:40-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160918 16:48:36< vultraz> which is why i said it might be worth widgetizing it
20160918 16:50:54-!- Greg-Boggs [~greg_bogg@c-76-115-139-154.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20160918 17:01:00-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 17:03:35-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Remote host closed the connection]
20160918 17:20:17< shadowm> loonycyborg: I told you everything I know.
20160918 17:21:05< shadowm> loonycyborg: Furthermore, if you arrived to the solution then you already know the problem.
20160918 17:21:41< shadowm> From the SO thread: "The problem is that the client may have no interest in the answer sometimes and exits early, so writing to that socket will cause a SIGPIPE and make my server crash."
20160918 17:22:15-!- Appleman1234 [~Appleman1@KD119104050103.au-net.ne.jp] has joined #wesnoth-dev
20160918 17:22:25< shadowm> Maybe I don't know what I'm talking about, but I think that's consistent with what I observed in the logs.
20160918 17:22:40< shadowm> So if you could fix this denial of service bug that'd be greatly appreciated.
20160918 17:23:43< celmin> vultraz: Huhwhat?
20160918 17:23:48-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 17:23:59< celmin> Note, I said "a bit hard" but it's certainly possible (with a cast).
20160918 17:25:03< shadowm> loonycyborg: And if you need extra information, tell me how to obtain it for you. I did not write the new network code, which is kind of the whole point, so I don't know how it works and as I've said a dozen times I don't have the energy to look at anything in Wesnoth that I'm not actively working on at the moment (i.e. 98% of it).
20160918 17:27:07-!- Netsplit *.net <-> *.split quits: Rhonda, vincent_c, oldlaptop
20160918 17:27:22< celmin> If the problem is a SIGPIPE, is there an issue with just installing a custom signal handler for it?
20160918 17:28:46< shadowm> Not at all considering that SDL_net was already doing that unconditionally as I said.
20160918 17:29:21-!- Netsplit over, joins: Rhonda, oldlaptop, vincent_c
20160918 17:32:42-!- irker094 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160918 17:38:51-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Remote host closed the connection]
20160918 17:39:40-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 17:46:14-!- prkc [~prkc@46.166.190.184] has quit [Ping timeout: 265 seconds]
20160918 17:56:36-!- prkc [~prkc@46.166.138.140] has joined #wesnoth-dev
20160918 18:05:04< bumbadadabum> celmin: daily poke
20160918 18:14:17-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Remote host closed the connection]
20160918 18:14:31-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 18:21:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 248 seconds]
20160918 18:39:50-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Remote host closed the connection]
20160918 18:41:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160918 18:44:42-!- Kwandulin [~Miranda@p200300760F2C71FC287E4782FE8F4B22.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160918 18:45:22-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 18:49:40-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has joined #wesnoth-dev
20160918 18:49:49-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Ping timeout: 255 seconds]
20160918 19:26:43-!- mjs-de [~mjs-de@x5ce4a284.dyn.telefonica.de] has joined #wesnoth-dev
20160918 19:33:59-!- boucman_work [~boucman@2a02-8428-034f-f800-9e32-0c7c-b391-6223.rev.sfr.net] has quit [Read error: Connection reset by peer]
20160918 19:34:09-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 19:38:25-!- irker634 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160918 19:38:25< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues a87898162861 / src/playsingle_controller.cpp: Fix bug: Map may not have starting positions https://github.com/wesnoth/wesnoth/commit/a878981628612317197e17ecfb2a09d8a0273ed5
20160918 19:38:27< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues fe08701b7ca3 / src/ (27 files in 13 dirs): Change most vector<bool> to dynamic_bitset https://github.com/wesnoth/wesnoth/commit/fe08701b7ca3639b17dca20023d2bb34433042c6
20160918 19:38:29< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues 8be428e9463d / changelog src/map/map.cpp: Honor special locations in terrain mask https://github.com/wesnoth/wesnoth/commit/8be428e9463daab3730550c79794c8af87ce19b1
20160918 19:38:31< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues b9a97d893d40 / changelog src/map/map.cpp: Merge pull request #775 from GregoryLundberg/GL_terrain_mask_positions https://github.com/wesnoth/wesnoth/commit/b9a97d893d40136e32ee3b508dddde9464ce9dc4
20160918 19:38:33< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 4cf5be674c6e / src/ (27 files in 13 dirs): Merge pull request #781 from wesnoth/dynamic_bitset https://github.com/wesnoth/wesnoth/commit/4cf5be674c6e2972c26a43a1d74315cf671f5f96
20160918 19:38:36< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 002f9a83c6b1 / src/gui/dialogs/end_credits.cpp: Yet again an attempt to appease travis https://github.com/wesnoth/wesnoth/commit/002f9a83c6b152244769707a4ffdac227c2b7872
20160918 19:38:39< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 183abfdabb61 / data/gui/window/game_load.cfg: Game Load: fixed layout with small number of saves https://github.com/wesnoth/wesnoth/commit/183abfdabb61b9eb1de9bdcb18957f813012f7a2
20160918 19:38:42< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 7eb13d218615 / src/gui/dialogs/end_credits.cpp: End Credits: removed the use of swap https://github.com/wesnoth/wesnoth/commit/7eb13d21861579a5040d6baf2e312e8b9b6d2a5f
20160918 19:38:45< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues fff2e3f19372 / src/gui/dialogs/end_credits.cpp: End Credits: fixed crash if exiting before scrolling starts https://github.com/wesnoth/wesnoth/commit/fff2e3f193721bcdab5b7b0c831323141acf19b1
20160918 19:38:48< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 7def4d57eb96 / src/gui/widgets/window.cpp: Removed some accidentally-committed debug code https://github.com/wesnoth/wesnoth/commit/7def4d57eb9699ff257fdfc95d6ae520b3be7462
20160918 19:38:51< irker634> wesnoth: loonycyborg wesnoth:fix-tunnel-issues b0fa8110f014 / src/ (campaign_server/campaign_server.cpp server/server_base.cpp): Convert remaining boost::bind to std::bind in campaignd code https://github.com/wesnoth/wesnoth/commit/b0fa8110f014a11ffbd81cb4d34afbb663a1f524
20160918 19:38:51< celmin> Ahhh, rebasing, okay.
20160918 19:38:53< irker634> wesnoth: loonycyborg wesnoth:fix-tunnel-issues 86dc5174a8bd / src/tools/exploder_utils.cpp: Fixed compile errors in exploder https://github.com/wesnoth/wesnoth/commit/86dc5174a8bd71267650274b401a6298b5c0b073
20160918 19:38:55< irker634> wesnoth: loonycyborg wesnoth:fix-tunnel-issues f821598ca3dc / src/tools/exploder_utils.cpp: Switch boost::shared_array to unique_ptr https://github.com/wesnoth/wesnoth/commit/f821598ca3dcf9818ff20965a197d0549ba0cbc8
20160918 19:38:57< irker634> wesnoth: Wedge009 wesnoth:fix-tunnel-issues 1a25d297b81d / changelog src/statistics.cpp: Don't duplicate recall count decrement when undoing a recall (bug #25060) https://github.com/wesnoth/wesnoth/commit/1a25d297b81dead1fd0e6e3c5b5da8d5caea6254
20160918 19:38:59< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues a477eb4014ba / src/statistics.cpp: Fixup 1a25d297b81d (unused variable) https://github.com/wesnoth/wesnoth/commit/a477eb4014bad07bfe5662982b2cbb7596975ded
20160918 19:39:02< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues ff9450eff297 / src/ (actions/move.cpp pathfind/teleport.cpp pathfind/teleport.hpp): Merge pull request #782 from wesnoth/fix-tunnel-issues https://github.com/wesnoth/wesnoth/commit/ff9450eff2975c02549f867c5fd8dc8dee000f0f
20160918 19:39:03< celmin> Was wondering why I suddenly got a ping when I hadn't pushed anything.
20160918 19:39:04< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues f4fcb330c81d / src/actions/move.cpp: Commented out unused parameter https://github.com/wesnoth/wesnoth/commit/f4fcb330c81d2c4090556671d5fcb6ce5630cbea
20160918 19:39:07< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 33e21fda4aaa / src/gui/widgets/generator.cpp: Commented out some aserts that were causing problems in a few places https://github.com/wesnoth/wesnoth/commit/33e21fda4aaa4d131bf57a83f66ee65a6fb20733
20160918 19:39:10< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues 076c327d565e / src/gui/widgets/generator.cpp: "Reactivate" removed asserts as a log message https://github.com/wesnoth/wesnoth/commit/076c327d565eea06cd8324d2239cd81b0ce1894b
20160918 19:39:12< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues bc7b27e49604 / data/gui/window/end_credits.cfg src/gui/dialogs/end_credits.cpp: Minor end credits tweaks https://github.com/wesnoth/wesnoth/commit/bc7b27e4960493849aa8cabc5238a0e66eb2f101
20160918 19:39:14< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 94b092fa9d37 / data/gui/widget/button_default.cfg: Added a definition for a really large button https://github.com/wesnoth/wesnoth/commit/94b092fa9d37f6aedd5f0fafd43abe0cf6a9c490
20160918 19:39:17< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues cf5640099614 / / (8 files in 4 dirs): MP Create: significant layout improvements https://github.com/wesnoth/wesnoth/commit/cf56400996142ca382ca57e26d11b3a6507658fd
20160918 19:39:20< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues bb6baf8a78f7 / src/game_preferences.cpp: Fixed bug where entering Preferences would clear saved password https://github.com/wesnoth/wesnoth/commit/bb6baf8a78f7a58a09205ed13e2bd7de687e4ece
20160918 19:39:23< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues b48d08d68751 / data/gui/window/drop_down_list.cfg: Drop Down List: some border fixes https://github.com/wesnoth/wesnoth/commit/b48d08d687515ff06e29e51fe48f845fcb6cd9ea
20160918 19:39:26< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 9682df6ad0c7 / data/gui/window/game_version.cfg: Game Version: small max height bump to avoid a scrollbar in tab 2 https://github.com/wesnoth/wesnoth/commit/9682df6ad0c70bc74f6a7d8761d769d0cbef1d16
20160918 19:39:29< irker634> wesnoth: Bär Halberkamp wesnoth:fix-tunnel-issues 6df17b37f481 / / (4 files in 3 dirs): EI: Rebalance S2 https://github.com/wesnoth/wesnoth/commit/6df17b37f481a72f6df65df0d87f1d589897af14
20160918 19:39:32< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues 898cf3b35d9c / data/campaigns/Heir_To_The_Throne/scenarios/ (4 files): HttT S22 Gryphons return https://github.com/wesnoth/wesnoth/commit/898cf3b35d9c63c2d01a3178f345e332f5e80bd2
20160918 19:39:34< irker634> wesnoth: Lari Nieminen wesnoth:fix-tunnel-issues 479e9062db1c / data/campaigns/Heir_To_The_Throne/scenarios/ (4 files): Merge pull request #737 from GregoryLundberg/GL_HttT_S22_gryphons_return https://github.com/wesnoth/wesnoth/commit/479e9062db1c24220ac3c6179507842df236bf65
20160918 19:39:36< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues fe85fa86cbf8 / src/playsingle_controller.cpp: Merge pull request #772 from GregoryLundberg/GL_allow_no_starting_positions https://github.com/wesnoth/wesnoth/commit/fe85fa86cbf8b6335000bf1bc6742391203f9165
20160918 19:39:38< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues d4ec02404827 / data/campaigns/Heir_To_The_Throne/scenarios/ (20b_Underground_Channels.cfg 25_HttT_Epilogue.cfg): HttT S20b Wose assistance quest https://github.com/wesnoth/wesnoth/commit/d4ec024048272aed0099ca8ca39d32b24219823a
20160918 19:39:40< irker634> wesnoth: Lari Nieminen wesnoth:fix-tunnel-issues f5b4ecf0c368 / data/campaigns/Heir_To_The_Throne/scenarios/ (20b_Underground_Channels.cfg 25_HttT_Epilogue.cfg): Merge pull request #736 from GregoryLundberg/GL_HttT_S20b_wose_assistance https://github.com/wesnoth/wesnoth/commit/f5b4ecf0c368b04b4d68e210a3e14ebeb1dc73be
20160918 19:39:42< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 553ac0fd0529 / src/ (53 files in 17 dirs): add wml_x/y() getter & setter in map_location https://github.com/wesnoth/wesnoth/commit/553ac0fd05290c5a3d3bbdb130e75f20d1c01bdb
20160918 19:39:44< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 3de50e526366 / src/ (game_preferences.cpp map/map.hpp): also mark terrains in map border as encountered. https://github.com/wesnoth/wesnoth/commit/3de50e526366848ff3ca9016c892f55c2a262bba
20160918 19:39:46< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues b43e5e697e84 / src/server/server.cpp: remove an include https://github.com/wesnoth/wesnoth/commit/b43e5e697e842527ae9faf01bfa8d07005c7e603
20160918 19:39:48< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues e8bf29c241a5 / src/ (13 files in 5 dirs): clean some includes https://github.com/wesnoth/wesnoth/commit/e8bf29c241a596d25aeb0ab35cc73aec80a1ac2c
20160918 19:39:50< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 050ef3ecd9a5 / src/ (8 files in 4 dirs): make gamemap.tiles_ non-jagged https://github.com/wesnoth/wesnoth/commit/050ef3ecd9a586408868e962efb57a54779fea5e
20160918 19:39:52< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 62bbab330d07 / src/ (editor/map/editor_map.hpp map/map.cpp map/map.hpp): remove unused code. https://github.com/wesnoth/wesnoth/commit/62bbab330d07a8d072d4296b571c8cc9088573f3
20160918 19:39:54< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 49621e3d2a7a / src/ (6 files in 4 dirs): remove border parameter from gamemap::overlay https://github.com/wesnoth/wesnoth/commit/49621e3d2a7ae8ca19c581c4880e7905e066a49f
20160918 19:39:56< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues d9a7729dbc1b / src/map/map.cpp: fix [terrain_mask][rule]use_old=yes https://github.com/wesnoth/wesnoth/commit/d9a7729dbc1b3ad1e0d82c7282f0eb416e4e5fcd
20160918 19:39:58< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 296faefa20ad / src/ (map/map.cpp map/map.hpp scripting/lua_common.cpp): remove gamemap.border_size_ https://github.com/wesnoth/wesnoth/commit/296faefa20ad023bfd578a4b1f1836d422d258f0
20160918 19:40:00< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 94611c696730 / src/map/ (map.cpp map.hpp): remove gamema::bordercache and gamema::terrainfrequencycache https://github.com/wesnoth/wesnoth/commit/94611c696730f11c7b177c325316bb6087245bd2
20160918 19:40:01-!- celmin [~celticmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The minstrel departs, to spread the music to the masses!]
20160918 19:40:02< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues e93c6ae75e1f / src/terrain/filter.cpp: make filter never pass on off-map locations https://github.com/wesnoth/wesnoth/commit/e93c6ae75e1fe1a7141fe536a2002ee071b0991d
20160918 19:40:04< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 96a8635ed989 / src/minimap.hpp: fix travis build https://github.com/wesnoth/wesnoth/commit/96a8635ed98923d8b9fa1169a0ed11a28d68fc4f
20160918 19:40:06< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 8aa373246c64 / src/map/map.cpp: cleanup gamemap::overlay https://github.com/wesnoth/wesnoth/commit/8aa373246c6411004980a0dc5783cdf3907948ab
20160918 19:40:08< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues 0caaaaa596cb / data/campaigns/Heir_To_The_Throne/scenarios/ (12_Northern_Winter.cfg 13_The_Dwarven_Doors.cfg): HttT S13 Add some random snow https://github.com/wesnoth/wesnoth/commit/0caaaaa596cb5e27f4a35a25991dc69fb052655b
20160918 19:40:10< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues 9bf3f9150367 / data/campaigns/Heir_To_The_Throne/scenarios/05b_Isle_of_the_Damned.cfg: HttT S05b Use an Advisor https://github.com/wesnoth/wesnoth/commit/9bf3f9150367c0e58e136a84d023afe58a4299bc
20160918 19:40:12< irker634> wesnoth: Lari Nieminen wesnoth:fix-tunnel-issues 1ab9b923dfa3 / data/campaigns/Heir_To_The_Throne/scenarios/ (12_Northern_Winter.cfg 13_The_Dwarven_Doors.cfg): Merge pull request #735 from GregoryLundberg/GL_HttT_S13_snow_ditritus https://github.com/wesnoth/wesnoth/commit/1ab9b923dfa3ffcecc8289a5e41b513b9890f99b
20160918 19:40:14< irker634> wesnoth: Lari Nieminen wesnoth:fix-tunnel-issues 6d36550e1e7e / data/campaigns/Heir_To_The_Throne/scenarios/05b_Isle_of_the_Damned.cfg: Merge pull request #731 from GregoryLundberg/GL_HttT_S05b_use_advisor https://github.com/wesnoth/wesnoth/commit/6d36550e1e7e0035028f609303e0e62901fffdbd
20160918 19:40:16< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues 946be0c02ca0 / / (5 files in 3 dirs): ~ADJUST_ALPHA() now takes formula https://github.com/wesnoth/wesnoth/commit/946be0c02ca09de4f0dabac0f2697dc42ae2f1a9
20160918 19:40:18< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues c9cbdb287dc3 / src/terrain/translation.hpp: Fix MSVC 2013 build https://github.com/wesnoth/wesnoth/commit/c9cbdb287dc3c158c5a17649d5053f583b180ea0
20160918 19:40:20< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues b84ec9f77552 / data/ai/micro_ais/mai-defs/ (animals.lua protect.lua): Micro AIs: fix attacks aspect setup when ca_id= key not given https://github.com/wesnoth/wesnoth/commit/b84ec9f7755213171d0bdac89b8f3780c3c017b3
20160918 19:40:22< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues 8ceac00cd3d8 / src/gui/widgets/settings.cpp: Finish support for GUI2 themes https://github.com/wesnoth/wesnoth/commit/8ceac00cd3d84d03fe7cade14563d01fa1e9aa03
20160918 19:40:24< irker634> wesnoth: Celtic Minstrel wesnoth:fix-tunnel-issues 31a972242a45 / src/ (log_windows.cpp log_windows.hpp wesnoth.cpp): When --wconsole causes the allocation of a new console, https://github.com/wesnoth/wesnoth/commit/31a972242a453be1b2f79043688eb3d07e4a7fac
20160918 19:40:26< irker634> wesnoth: Charles Dang wesnoth:fix-tunnel-issues 33653e396945 / data/gui/ (widget/chatbox.cfg window/lobby_main.cfg): Reduced chatbox size slightly https://github.com/wesnoth/wesnoth/commit/33653e396945bf0e5d0ca56565149ae1f3328ba3
20160918 19:40:29< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues 304cb0dd1aba / src/pathfind/teleport.cpp: [tunnel] tag: change default for pass_allied_units key to yes https://github.com/wesnoth/wesnoth/commit/304cb0dd1abad7b4a3d49fc91c51e1309adb8d30
20160918 19:40:31< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 60d114b606b3 / / (8 files in 6 dirs): finish moving cavegen (cave_map_generator.cpp)  to lua. https://github.com/wesnoth/wesnoth/commit/60d114b606b305099c690a8c9d6cc0eaff9a905c
20160918 19:40:33< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues 55edd99c3eb7 / src/game_initialization/flg_manager.cpp: attempt to fix wrong recruitlist in UtBS 1 https://github.com/wesnoth/wesnoth/commit/55edd99c3eb7a471b33cfee36f7102d38679a728
20160918 19:40:35< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues 0e87d7fd4365 / data/ai/lua/ca_high_xp_attack.lua: High XP attack CA: respect the default 'avoid' aspect https://github.com/wesnoth/wesnoth/commit/0e87d7fd436588f88fc093fd460b80903e1c5f4e
20160918 19:40:37< irker634> wesnoth: Gregory A Lundberg wesnoth:fix-tunnel-issues fc4c75e56d65 / data/campaigns/Heir_To_The_Throne/scenarios/23_Test_of_the_Clans.cfg: HttT S23 Fix bug: Unclear objective https://github.com/wesnoth/wesnoth/commit/fc4c75e56d6558689e56a38560d36c9342a74bb2
20160918 19:40:39< irker634> wesnoth: gfgtdf wesnoth:fix-tunnel-issues f2635e5ad65d / changelog: Update changelog https://github.com/wesnoth/wesnoth/commit/f2635e5ad65d991b8911fac1cfc60f88836c3f49
20160918 19:40:41< irker634> wesnoth: Lari Nieminen wesnoth:fix-tunnel-issues 8397201f60a7 / data/campaigns/Heir_To_The_Throne/scenarios/23_Test_of_the_Clans.cfg: Merge pull request #789 from GregoryLundberg/GL_HttT_S23_objectives https://github.com/wesnoth/wesnoth/commit/8397201f60a7ceb2cce2160b2e306715bea8ff78
20160918 19:40:43< irker634> wesnoth: doofus-01 wesnoth:fix-tunnel-issues 376d45a292fd / / (48 files in 4 dirs): new terrain: Merfolk or aquatic camp (#773) https://github.com/wesnoth/wesnoth/commit/376d45a292fdf2bd24802bb5d6de636189d481f6
20160918 19:40:45< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues 6f98888e246d / src/pathfind/ (pathfind.cpp teleport.cpp teleport.hpp): Tunnels: fix vision not being propagated through tunnels https://github.com/wesnoth/wesnoth/commit/6f98888e246d0f13fe71e788e0c973169c2c595b
20160918 19:40:47< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues c505c962a28a / : Merge remote-tracking branch 'origin/fix-tunnel-issues' into fix-tunnel-issues https://github.com/wesnoth/wesnoth/commit/c505c962a28ae5faa5ee46d48f1029aa4d1cfcc8
20160918 19:40:54< mattsc> Ugh.
20160918 19:40:58< mattsc> Sorry.
20160918 19:41:02< mattsc> Yes, rebasing ...
20160918 19:41:16< mattsc> I get the feeling that I should really work on a fork ...
20160918 19:42:49< mattsc> The thing though is, I thought I had squashed those commits.  And there they are back in the PR.
20160918 19:43:18< JyrkiVesterinen> When you pushed to the wesnoth repository, did you use the --force switch?
20160918 19:43:42-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160918 19:43:45< JyrkiVesterinen> I recall that Git doesn't automatically suggest --force when you try to push after rebasing.
20160918 19:43:58< JyrkiVesterinen> If you merge instead, you end up with duplicate commits.
20160918 19:44:20< mattsc> Yeah, that’s probably it …
20160918 19:44:41< JyrkiVesterinen> Sigh. It's a major design mistake in Git in my opinion.
20160918 19:44:45< celticminstrel> Ah, yeah, that was a merge not a rebase.
20160918 19:44:54< celticminstrel> Which is a design mistake exactly?
20160918 19:44:58< celticminstrel> Duplicate commits?
20160918 19:45:23< JyrkiVesterinen> Git basically encourages newcomers to make the "rebase, then merge" mistake.
20160918 19:45:30< mattsc> So, if I squash again, and the force push, will this channel be flooded with another 64 commit messages?
20160918 19:45:31< iceiceice> JyrkiVesterinen, usually people don't want to automatically force push
20160918 19:45:34< iceiceice> because it destroys history
20160918 19:45:45-!- Kwandulin [~Miranda@p200300760F2C71FC287E4782FE8F4B22.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160918 19:45:49< JyrkiVesterinen> Yes, I know. --force has its dangers.
20160918 19:45:51< celticminstrel> mattsc: Squashing seems irrelevant.
20160918 19:46:40< celticminstrel> mattsc: How many commits were there supposed to be in the PR, exactly?
20160918 19:46:40< mattsc> celticminstrel: ?
20160918 19:46:48< mattsc> At this point, one.
20160918 19:47:22< iceiceice> JyrkiVesterinen, i dont think its a design mistake, i think rebase is just thoguht of as an advanced feature
20160918 19:47:37< tad_> When I work on a private branch preparing a PR, I force and rebase interactively and whatever because, to me, history does not matter on personal, private work. Once it goes into the master, however, history is important and should not be changed without careful throught
20160918 19:47:43< celticminstrel> I think if you rebase ('git rebase master') and force-push, we'll only see the one post here.
20160918 19:47:50< iceiceice> if someone doesn't know what they are doing, it's better to merge every time i think
20160918 19:47:53< tad_> mattsc: ALWAYS work on a branch!
20160918 19:47:58< celticminstrel> Because all the other commits will then already exist in the github repo.
20160918 19:48:01< JyrkiVesterinen> iceiceice: And yet rebase is needed all the time.
20160918 19:48:03< celticminstrel> tad_: Pretty sure he's working on a branch.
20160918 19:48:12< JyrkiVesterinen> You work on a feature and create a pull request.
20160918 19:48:23< iceiceice> no, it depends what work flow you are using
20160918 19:48:24< mattsc> tad_: it is a branch, but it’s a branch off master, not a fork
20160918 19:48:27< iceiceice> in some projects you always merge
20160918 19:48:28< JyrkiVesterinen> Then master gets updated and you need to update the PR by rebasing.
20160918 19:48:34< iceiceice> you can simply merge instead
20160918 19:48:37< celticminstrel> I read something somewhere saying that rebasing is better for resolving conflicts.
20160918 19:48:47< JyrkiVesterinen> And you need to update your fork in GitHub, i.e. push.
20160918 19:49:08< JyrkiVesterinen> And that push requires --force, but Git doesn't advice you to use it. It tells you to merge instead.
20160918 19:49:11< celticminstrel> At least, I think that's what it said. Can't quite remember.
20160918 19:49:16< iceiceice> no, it only requires force if you destoryed history.
20160918 19:49:20< iceiceice> if you always merge, it will work
20160918 19:49:32< iceiceice> i'm not advocating that,
20160918 19:49:54< iceiceice> i mean you can decide for yourself if you like having mergecommits in the history
20160918 19:49:58< celticminstrel> It was a post about disadvantages of the "merge workflow".
20160918 19:50:02< JyrkiVesterinen> At least I recall that Git almost always needs a force-push after rebase.
20160918 19:50:03< iceiceice> but it is a valid option
20160918 19:50:18< iceiceice> yeah, rebase is not strictly necessary though.
20160918 19:50:23< celticminstrel> I don't mind having merge commits, but I like to avoid commits merging master into a branch (if it's a huge branch I can make an exception).
20160918 19:50:45< iceiceice> its just kind of an advanced feature if you want to have precise control over how the history will look
20160918 19:52:00< iceiceice> celticminstrel, yeah i think the cases when rebase works well, it's better
20160918 19:52:11< iceiceice> idk
20160918 19:52:30< celticminstrel> My impression was that Jyrki was partly complaining about the message git gives you when a push is rejected.
20160918 19:52:44< JyrkiVesterinen> Yes, that was my main point.
20160918 19:53:03< celticminstrel> The message suggests that you merge but does not mention that you could instead rebase.
20160918 19:53:40< iceiceice> i thought the suggestion was that, the message should tell you you should force push
20160918 19:53:43< iceiceice> because you did rebase
20160918 19:53:57-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 19:53:58< travis-ci> spixi/wesnoth#15 (chart_engine - 1dc074b : Spixi): The build has errored.
20160918 19:53:59< travis-ci> Build details : https://travis-ci.org/spixi/wesnoth/builds/160876168
20160918 19:53:59-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 19:54:00< tad_> Working off a personal fork has the advantage that you can catch mistakes like just occurred and correct them before merging them to master.
20160918 19:54:26< iceiceice> but i think actually thats likely to just cause a lot of newbies to force push when they shouldnt
20160918 19:54:33< JyrkiVesterinen> Well, fixing mistakes is possible in upstream branches too. Matt's changes are not yet in master.
20160918 19:54:50< mattsc> tad_: 20160918 19:41:16< mattsc> I get the feeling that I should really work on a fork ...
20160918 19:55:00< iceiceice> JyrkiVesterinen, they are in the main wesnoth repo though
20160918 19:55:01< mattsc> tad_: yes, that’s what that meant 
20160918 19:55:07 * tad_ nods.
20160918 19:55:10< iceiceice> anyone who cloned and has that branch would get an error
20160918 19:55:17< iceiceice> if it is rebased and force pushed
20160918 19:55:21< iceiceice> and they try to pull later
20160918 19:55:32< iceiceice> doesn't really matter if its master or not
20160918 19:55:58< JyrkiVesterinen> I think it does matter. Other developers rarely check out branches other than master.
20160918 19:56:12< iceiceice> it doesn't matter if they checked it out or not
20160918 19:56:19< JyrkiVesterinen> I othen force-push to my branches in the main repository, in fact.
20160918 19:56:21< iceiceice> if you do git clone iirc you get most branches
20160918 19:56:44< iceiceice> i dont remember if you get all tags... owuld have to look at it again
20160918 19:56:52< iceiceice> they might not often *update* all those branches
20160918 19:57:44< JyrkiVesterinen> AFAIK, cloning doesn't give you branches other than master.
20160918 19:58:03< JyrkiVesterinen> If I run "git branch" in my local repository, I only get master and my local branches.
20160918 19:58:13< irker634> wesnoth: Celtic Minstrel wesnoth:master 1816e1147f55 / / (19 files in 10 dirs): Remove asserts.hpp https://github.com/wesnoth/wesnoth/commit/1816e1147f553c075b73c668117fb9833c5987c6
20160918 19:58:15< irker634> wesnoth: Celtic Minstrel wesnoth:master a63903cca047 / src/gui/ (5 files in 2 dirs): Use FAIL() instead of VALIDATE(false) https://github.com/wesnoth/wesnoth/commit/a63903cca047fe179adaa632422d98042377f912
20160918 19:59:20< celticminstrel> I'm not quite sure, but I think a pull force-updates remote-tracking branches, doesn't it?
20160918 19:59:55< mattsc> Is it to late to claim that I meant to merge, since that’s what they suggest you do on some Interweb channels?  ;)
20160918 20:00:32< mattsc> Anyways, concrete question: do I need to do something about the commits in the PR: https://github.com/wesnoth/wesnoth/pull/790
20160918 20:00:52< JyrkiVesterinen> celticminstrel: I don't know. When I pull from the main repository, I explicitly tell it to only pull master.
20160918 20:00:59< JyrkiVesterinen> $ git pull wesnoth master
20160918 20:01:09< JyrkiVesterinen> (I have named the main remote "wesnoth")
20160918 20:01:55< celticminstrel> JyrkiVesterinen: Yeah, I set my remote to not track non-master branches too.
20160918 20:02:02< celticminstrel> Oh, wait, that's different.
20160918 20:02:31< celticminstrel> I mean if I do "git branch -r" I don't see the remote-tracking branch for every branch in the main repo.
20160918 20:02:51< mattsc> And I think the next thing is to walk away and stop meddling with things I don’t understand.  :P
20160918 20:03:18< celticminstrel> mattsc: What I would try is 'git rebase master' and then check the log to see if the duplicate commits are still there (they probably won't be).
20160918 20:05:09< mattsc> celticminstrel: and then I spam this channel with another 6 dozen commits?
20160918 20:05:34< celticminstrel> If the duplicate commits are not in your log, then a force-push should not spam the channel.
20160918 20:05:54< JyrkiVesterinen> My experience with irker is that it only announces commits which aren't in master.
20160918 20:06:47< JyrkiVesterinen> See yesterday's log (at around 8:00 UTC): http://wesnoth.org/irclogs/2016/09/%23wesnoth-dev.2016-09-17.log
20160918 20:06:57< celticminstrel> I was assuming it only announced commits that aren't already in the repo in some form...
20160918 20:07:17< celticminstrel> It's nice that that's properly percent-escaped.
20160918 20:07:27< JyrkiVesterinen> I did three force-pushes in rapid succession. Irker announced one, four, and three commits respectively.
20160918 20:07:35< mattsc> Hmm
20160918 20:07:42< mattsc> Okay.
20160918 20:08:47< mattsc> So after ‘git rebase master’ it tells me that my branch and origin/fix-tunnel-issues have diverged, with 3 commits for each
20160918 20:09:04< mattsc> * 3 different commits
20160918 20:09:13< mattsc> It also tells me to pull.
20160918 20:09:36< mattsc> The log only shows one of commit, as celticminstrel said
20160918 20:09:45< JyrkiVesterinen> Do NOT pull. Force-push instead.
20160918 20:09:57< JyrkiVesterinen> Pulling would require you to merge and create duplicate commits.
20160918 20:10:25< JyrkiVesterinen> This is exactly the trap that Git gives you after rebasing. The one I complained about above.
20160918 20:10:44< mattsc> Right, and that’s what I did before.
20160918 20:10:53< tad_> mattsc: The hashes diverged. Three down: old version. Three up: new version. Even if the new and old are the same. All git knows is the hashes differ
20160918 20:11:00< mattsc> So, simply ‘git push -f’ now?
20160918 20:11:05< JyrkiVesterinen> Yes.
20160918 20:11:40< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues eb6cea678ed3 / src/pathfind/ (pathfind.cpp teleport.cpp teleport.hpp): Tunnels: fix vision not being propagated through tunnels https://github.com/wesnoth/wesnoth/commit/eb6cea678ed3b4f77b2b54c955b0988835862854
20160918 20:11:47< JyrkiVesterinen> :)
20160918 20:12:22< celticminstrel> That's because the commit's parent is part of the hash.
20160918 20:12:31 * tad_ nods.
20160918 20:12:50< mattsc> Everybody, thanks and sorry for all the stupid questions!
20160918 20:13:12< celticminstrel> If you made a commit, then erased it and made the exact same commit again, it would probably have the same hash. (Unless the date and time is also part of the hash.)
20160918 20:13:13< JyrkiVesterinen> The questions aren't stupid. What is stupid is the design of Git rebase and push.
20160918 20:13:29< tad_> No such thing as stupid question. And you learn best when you make errors.
20160918 20:13:45< celticminstrel> Okay, now I need to figure out how the command executor can access the play controller.
20160918 20:13:54< mattsc> Well, yeah, I’ve used all of those lines before myself.  ;)  I prefer making the mistakes not in a public forum though.
20160918 20:15:17< celticminstrel> I'm pretty sure it's actually impossible.
20160918 20:15:33< celticminstrel> Hmm...
20160918 20:15:41< tad_> Public embarrassment is an excellent motivator to learn, though.
20160918 20:15:47< celticminstrel> Maybe if I made show_menu return a string specifying the submenu to show?
20160918 20:15:53< mattsc> Hehe, yeah.
20160918 20:16:23< tad_> After a few decades you sorta learn to accept the embarrassment and move on ...
20160918 20:16:35< mattsc> I think I did pretty much everything correctly up to the ‘pull’ command.  For some reason I had gotten the impression that a normal ‘pull’ should not be possible in that situation and wanted to see what it would tell me if I did.  And thus fell into the trap JyrkiVesterinen was talking about.
20160918 20:17:16< mattsc> tad_: yeah, I have a couple decades of that under my belt too (albeit not with shared software development), so I am not too embarrassed.
20160918 20:17:37< mattsc> I’m not one of the young guys here either, although you still have a few years on me.  :)
20160918 20:17:44< celticminstrel> Should the show_menu function really be responsible also for executing the selected menuitem?
20160918 20:18:51< tad_> Well, you haven't lived until you push a commit to release without noticing the typo and end up spending a week working with CERT-CC drafting your statement of public embarrassment/
20160918 20:19:27< mattsc> Yeah, that sounds like fun …
20160918 20:22:49< JyrkiVesterinen> tad_: I have one somewhat similar experience. An update to the game Angry Birds Slingshot Stella introduced a new game mode. Too bad that game mode had a bug that if the player didn't play it for a week, it became inaccessible.
20160918 20:23:12< JyrkiVesterinen> QA failed to notice it because "don't play the game for a week" is not exactly a common test.
20160918 20:23:28< JyrkiVesterinen> We pushed out a hotfix ASAP, of course.
20160918 20:23:32-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 20:23:33< travis-ci> wesnoth/wesnoth#11059 (fix-tunnel-issues - c505c96 : mattsc): The build has errored.
20160918 20:23:33< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160878393
20160918 20:23:33-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 20:23:33-!- iceiceice [~chris@pool-71-172-187-9.nwrknj.east.verizon.net] has quit [Ping timeout: 240 seconds]
20160918 20:23:42< JyrkiVesterinen> (I was the one who *found* the bug, I didn't cause it.)
20160918 20:23:53< celticminstrel> I can't help but wonder how that would happen.
20160918 20:24:30< tad_> JyrkiVesterinen: Mine was an = instead of == with passed testing, went live, resulted in a worm and having 35M users and about 2M servers needing a security patch last week.
20160918 20:24:47< JyrkiVesterinen> The game mode consisted of events, and there was initially an idea to have time limits in them. Some code was written to expire each event after a week.
20160918 20:25:00< celticminstrel> Why does XCode give mismatched exception specification errors when a declaration has unrelated syntax errors...
20160918 20:25:06< JyrkiVesterinen> The feature was forgotten, but the code remained and rotted.
20160918 20:25:23< JyrkiVesterinen> When any even expired, it went into an infinite loop.
20160918 20:25:26< JyrkiVesterinen> *event
20160918 20:25:38< tad_> Shit happens. You deal with it and move on.
20160918 20:26:01< celticminstrel> Speaking about = vs ==, I think there's still one outstanding warning about that in MSVC.
20160918 20:28:35< gfgtdf> tad_: you said in the changeloc that its possible that multiple spaceial locatiosn can be at the same place.
20160918 20:29:14< gfgtdf> tad_: but afaik even though the gamemap clas supports that the map reading/writing code doesn't
20160918 20:29:58< gfgtdf> tad_: so that when you reload a game (whcih impled serializing and unserialzing the map data) all but one spacial locations are lost for each place
20160918 20:33:06< celticminstrel> gfgtdf: Sounds like something easily fixed?
20160918 20:33:33< tad_> gfgtdf: Hmm. Seems we have a disconnect between what can be done while running vs the save/load logic. One or the other needs fixing.
20160918 20:33:45< celticminstrel> Actually, can bimap have duplicates like that?
20160918 20:34:08< gfgtdf> celticminstrel: in this case tey can
20160918 20:34:24< tad_> celticminstrel: Seemed to work. 
20160918 20:34:35-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Remote host closed the connection]
20160918 20:34:52< tad_> Easy fix is to add a special check to enforce unique on both sides of the map.
20160918 20:34:57< gfgtdf> tad_: whne i fist implemented it, my orinal plan was to allow multiple location ids on the same place, but then i didnt know how to integrate it in the editor so i left it for now.
20160918 20:35:30< tad_> gfgtdf: To do it, you need multiple masks layered over the location.
20160918 20:35:33< celticminstrel> Implementing it in save/load would be easy though.
20160918 20:35:41< celticminstrel> In the editor, I'm not sure.
20160918 20:36:10< gfgtdf> tad_: or by editong the map data manually (in cae we fix the issue).
20160918 20:36:11< tad_> celticminstrel: Can you load the map from a saved game into the editor?
20160918 20:36:39< gfgtdf> tad_: well you can sureley copy the map_data from it to a new file and open that in the editor
20160918 20:36:48< celticminstrel> I'm not sure.
20160918 20:36:53< celticminstrel> Maybe?
20160918 20:37:12< celticminstrel> I think there's an option to load a game and then export just the map.
20160918 20:37:16< celticminstrel> But that's not in the editor.
20160918 20:37:25 * tad_ nods.
20160918 20:37:29< tad_> Never tried that.
20160918 20:38:45-!- ChipmunkV [~vova@static-89-94-113-91.axione.abo.bbox.fr] has quit [Quit: ChipmunkV]
20160918 20:38:48< tad_> Anyway .. if we fix it so only one name per location .. we have to answer the question of what happens to "1" and (20,20) when a mask wants to put "1" at (20,21) and "Advisor" at (20,20) .. 
20160918 20:39:29< tad_> Note it works now becuase the delete-and-add for "1" happens after the add for "Advisor" but the duplicate is allowed for the moment between the two.
20160918 20:40:35< tad_> So what happens if the mask does not move "1" elsewhere? Do we reject the add for "Advisor" or imply a delete for side "1" starting position
20160918 20:41:47< gfgtdf> hmm i actuallt think its best if we just update the eadong/writing code ot allow multiple location ids at one place.
20160918 20:43:18< tad_> I agree.  If that means editing the map is a problem, well, it's not happening now so we have time to figure out how the Map Editor should stack flags and handle deleting one from a stack.
20160918 20:43:50-!- JyrkiVesterinen [~JyrkiVest@78-27-104-98.bb.dnainternet.fi] has quit [Quit: Going to bed]
20160918 20:44:37< celticminstrel> The map editor should preserve multiple location IDs if possible though.
20160918 20:44:51< celticminstrel> ie, if you load a map and tweak a few terrains and save, all location IDs should be untouched.
20160918 20:45:15< tad_> As a first step, yes. So long as you don't tweak the flags.
20160918 20:46:07< tad_> bimap does not seem to care that A or B are unique, only that the pair (A,B) is.
20160918 20:46:27< celticminstrel> Ah, that kinda makes sense.
20160918 20:46:29< tad_> but I could be wrong.
20160918 20:46:46< gfgtdf> tad_: no this actuall dpends on how you configure the bimap
20160918 20:46:48< celticminstrel> Does it also enforce constness of A and B?
20160918 20:46:56< gfgtdf> tad_: that is on what template parmaters you give to it
20160918 20:46:58< tad_> Boost documentation is about as informative as reading /dev/null
20160918 20:47:00< celticminstrel> gfgtdf: Ehhh what.
20160918 20:47:05< celticminstrel> tad_: Haha, yeah.
20160918 20:47:13< celticminstrel> Mind you, it does depend on the library to some extent.
20160918 20:47:45< tad_> Well, when doing Boost across versions, maybe reading /dev/urandom
20160918 20:48:02< celticminstrel> I found the Spirit documentation to be pretty good when I first looked at it. (That was before the rewrote the whole library and rendered it incomprehensible, which is why I ended up using Spirit Classic instead of the new Spirit when I needed it.)
20160918 20:48:15< celticminstrel> ^before they3
20160918 20:53:28< tad_> So I'm reading the password handling code for the campaign server and it's hashing. Good. Is it hashing correctly? Check Boost. 404 Not Found. Huh? SO has the answer, not Boost: the function was deprecated years ago. It and its header are still there, but 404 the documentation.
20160918 20:54:09-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 20:54:10< travis-ci> wesnoth/wesnoth#11060 (master - a63903c : Celtic Minstrel): The build was broken.
20160918 20:54:10< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160881858
20160918 20:54:10-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 20:54:25< tad_> Broken?
20160918 20:57:39< celticminstrel> Guess so... those last two commits were cherry-pikcked from my branch, so maybe I messed up on the conflict resolution...
20160918 20:57:55< celticminstrel> Gah, I pressed delete, why is that K still there.
20160918 20:58:19< tad_> gfgtdf: If the save-game load for the map uses the common map load function, and the editor uses that same function (both: probably true) then we'll need to add a specific check for a duplicate name on load.
20160918 20:59:02< tad_> gfgtdf: At present, the change I made assumes no duplicate names on the individual maps and only handles ensuring no duplicates when merging them
20160918 20:59:10< celticminstrel> Anyone care to tell me what the error is? I suppose I could type out the URL in Firefox...
20160918 20:59:52-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 255 seconds]
20160918 21:00:11< celticminstrel> If it's "unknown identifier ERROR_LOG" I know exactly how to fix it and can do so through the website.
20160918 21:00:16< tad_> OH my. I'll get it of my local console
20160918 21:00:28< celticminstrel> Hm?
20160918 21:00:40< celticminstrel> I guess I need to log into github on that computer though.
20160918 21:00:43< tad_> First one: src/gui/widgets/settings.cpp:639:58: error: no match for ‘operator<<’ (operand types are ‘const char [42]’ and ‘const string {aka const std::__cxx11::basic_string<char>}’)     ERROR_LOG("default definition not found for control " << control_type);
20160918 21:01:04< celticminstrel> Yeah, that's the one; I got it after rebasing the branch (from which I cherry-picked the commits) onto master.
20160918 21:01:40< tad_> Thre's a lot more. Can't tell if they're a result of that one or different. 
20160918 21:01:48< celticminstrel> Same file?
20160918 21:02:39< tad_> Looks like the spam on my screen is all a result of that one error.
20160918 21:03:07-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160918 21:04:55< irker634> wesnoth: Celtic Minstrel wesnoth:master 713711a2ccac / src/gui/widgets/settings.cpp: Fixup botched conflict resolution https://github.com/wesnoth/wesnoth/commit/713711a2ccacf00d6910ca180b9f8928b651d679
20160918 21:08:43< tad_> MAde it past settings.cpp so that's good
20160918 21:08:46< celticminstrel> Though I suspect that one actually wasn't a conflict and that's why it got missed...
20160918 21:11:50< celticminstrel> I really need to turn off MSVC's brace autocompletion.
20160918 21:20:18< tad_> All those files out-of-date because a header they probably don't need but which came along because something they used needed something which needed something which ...
20160918 21:23:20< celticminstrel> Yeah...
20160918 21:23:57< celticminstrel> I did improve that a bit awhile back - forward-declaring the config class instead of including config.hpp wherever possible.
20160918 21:24:07< celticminstrel> But I'm sure it could still use quite a bit more improvement.
20160918 21:24:41< tad_> Encapsulation would help a lot. But there's a lot to refactor for that to make a significant difference.
20160918 21:26:39< celticminstrel> I thought that windows received draw events so that they'd always update when needed, but it seems like that's not working...
20160918 21:26:44< celticminstrel> Maybe because...
20160918 21:27:08< celticminstrel> game_display::draw() is called after events::pump()?
20160918 21:29:21-!- mjs-de [~mjs-de@x5ce4a284.dyn.telefonica.de] has quit [Remote host closed the connection]
20160918 21:30:15-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 21:30:16< travis-ci> wesnoth/wesnoth#11062 (fix-tunnel-issues - eb6cea6 : mattsc): The build is still failing.
20160918 21:30:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160883708
20160918 21:30:16-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 21:31:22< tad_> ^ legacy. s/b fixed
20160918 21:34:21< Aginor> 09:27 < celticminstrel> game_display::draw() is called after events::pump()?
20160918 21:34:30< Aginor> ^ I think that's how it should be
20160918 21:34:48-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has joined #wesnoth-dev
20160918 21:34:59< Aginor> I maintain that we want to 1. Process inputs, 2. Update state, 3. Render
20160918 21:35:40< celticminstrel> Aginor: Makes sense, but if I understand correctly, GUI2 elements redraw in response to a DRAW event, don't they?
20160918 21:35:58< Aginor> celticminstrel: both GUI1 and GUI2 does that
20160918 21:36:14< Aginor> it's something I think should change
20160918 21:36:35< Aginor> I was experimenting with cleaning this up in the renderpath_redo branch
20160918 21:36:57< Aginor> but it turned into a mess and too many other things changed at the same time for it to be entirely viable
20160918 21:37:42< irker634> wesnoth: Celtic Minstrel wesnoth:gui2_floating_textbox 916f8f63caf4 / src/ (9 files in 2 dirs): Use GUI2 for floating textboxes (WIP) https://github.com/wesnoth/wesnoth/commit/916f8f63caf44f881661f5797264ac3d37d1a830
20160918 21:37:44< irker634> wesnoth: Celtic Minstrel wesnoth:gui2_floating_textbox 6d8e44a4f0db / / (9 files in 5 dirs): Try floating textbox as popup instead of a standalone widget https://github.com/wesnoth/wesnoth/commit/6d8e44a4f0dbb9d756542feb37c1bf662065bc90
20160918 21:37:52< celticminstrel> Well, I've pushed what I got so far so people can look at it...
20160918 21:38:02< celticminstrel> It doesn't really work at all, though. :/
20160918 21:38:14< celticminstrel> The textbox draws once, then disappears once you scroll.
20160918 21:38:24< Aginor> scroll what?
20160918 21:38:29< celticminstrel> The map.
20160918 21:38:32< Aginor> ah
20160918 21:38:43< Aginor> mark them as dirty on a scroll event
20160918 21:38:54< tad_> Like the error box for recall which goes away so quickly you can't read it, and really never even see it?
20160918 21:38:57< celticminstrel> I'm not sure if it responds to any events, but it does prevent the main UI from receiving regular key/mouse events somehow.
20160918 21:39:03< celticminstrel> tad_: Huh?
20160918 21:39:20< celticminstrel> Aginor: You mean, mark the textbox as dirty on a scroll event?
20160918 21:39:37< tad_> Right-click to recall when the list is empty. Should show an error box. But, if so, it's so fast you never see it.
20160918 21:39:39< Aginor> celticminstrel: that's because gui2 runs in a separate event context, so it'll always block gui1 components from receiving any events while active
20160918 21:39:48< Aginor> celticminstrel: yes
20160918 21:39:58< tad_> gtg bbl
20160918 21:40:02-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has quit [Quit: Page closed]
20160918 21:40:10< Aginor> celticminstrel: you'll also have move it on the scroll as well, obviously ;)
20160918 21:40:47< celticminstrel> Aginor: I don't think these are even GUI1 components exactly...? I thought the play controller or something had an event context.
20160918 21:40:55< celticminstrel> Aginor: Huh? Why do I need to move it?
20160918 21:41:34< Aginor> celticminstrel: the map labels are supposed to be tied to a hex, so if the hex moves position you move it
20160918 21:41:43< Aginor> celticminstrel: they are effectively gui1 components
20160918 21:42:04< celticminstrel> Ah. Well, this isn't a map label, it's a GUI2 popup window (sort of similar to GUI2 tooltips).
20160918 21:42:35< Aginor> right
20160918 21:42:43< celticminstrel> If I switch away from Wesnoth and back, it does redraw properly after scrolling (I just discovered this).
20160918 21:42:59< celticminstrel> Except it doesn't redraw over water.
20160918 21:43:06< Aginor> that's because I added code to redraw everything on those window events
20160918 21:43:38< celticminstrel> It doesn't seem to be covered by events::raise_draw_event()
20160918 21:43:51< celticminstrel> (Which is being called immediately after events::pump())
20160918 21:44:07< Aginor> events::raise_draw_event() is kind of pointless and should be removed in the long run
20160918 21:44:29< Aginor> it'd be much better to separate the event handling from the drawing
20160918 21:44:42< Aginor> and lead to a cleaner separation from state and rendering
20160918 21:44:52< Aginor> s/from/between/
20160918 21:45:46< celticminstrel> I thought you were the one who added raise_draw_event... am I imagining things again?
20160918 21:46:10< Aginor> I most certainly didn't ;)
20160918 21:46:34< Aginor> I did the draw_all stuff, to do a full redraw of every event context
20160918 21:46:53< celticminstrel> Hmm. Since it does draw if I switch away and back, maybe I could somehow put the drawing issue on hold and work on the events trapping issue...
20160918 21:46:57< celticminstrel> Somehow...
20160918 21:47:08< Aginor> but I think I need to take this further and add a render manager class
20160918 21:47:14< celticminstrel> Tooltips don't trap events, but... that's because they work in the GUI2 event chain...
20160918 21:47:41< celticminstrel> I don't know how the older tooltips work.
20160918 21:47:51< celticminstrel> Probably nothing so fancy.
20160918 21:48:12< Aginor> I don't think they're gui2 though, not there at least
20160918 21:48:17< celticminstrel> Yeah.
20160918 21:48:25< celticminstrel> That's why I said "older tooltips".
20160918 21:48:28< Aginor> indeed
20160918 21:48:51< Aginor> you could try adding gui2 to the same event context as gui1, but it'll turn rather interesting at that point too
20160918 21:49:31< Aginor> or just add a better event handling framework in general and make sure it can emulate the current behavior except for where it explicitly doesn't
20160918 21:49:42< celticminstrel> Bleh.
20160918 21:49:46< Aginor> but that's all a lot of work
20160918 21:49:53< celticminstrel> I'm not sure how I'd even get started on something like that.
20160918 21:49:59< Aginor> probably not that much code, but a lot of work
20160918 21:50:09< Aginor> celticminstrel: you draw on a whiteboard or piece of paper
20160918 21:50:15< Aginor> celticminstrel: and talk about it a lot ;)
20160918 21:50:19< celticminstrel> Adding GUI2 to the same event context seems like an idea, as long as it can be done only for floating textboxes.
20160918 21:50:27< Aginor> nope
20160918 21:50:33< Aginor> it's all or nothing with gui2
20160918 21:50:42< Aginor> it does its own internal distributing
20160918 21:50:51< Aginor> hmm
20160918 21:50:54< celticminstrel> I'm not entirely clear on how all this event context stuff works.
20160918 21:50:59< Aginor> tie gui1 into the gui2 event loop
20160918 21:51:04< Aginor> could work
20160918 21:51:21< celticminstrel> So, make GUI1 use the tdistributor / tdispatcher system?
20160918 21:51:25< Aginor> I better disappear again, this is too distracting ;)
20160918 21:51:26< Aginor> yeah
20160918 21:51:30< celticminstrel> Although GUI1 is almost dead, so is it worth it? :/
20160918 21:51:53-!- Appleman1234 [~Appleman1@KD119104050103.au-net.ne.jp] has quit [Ping timeout: 240 seconds]
20160918 21:51:56< Aginor> until you move the game display to gui2, it's far from dead
20160918 21:52:07< Aginor> that's the most significant aspect of it
20160918 21:52:18-!- travis-ci [~travis-ci@ec2-54-158-59-140.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 21:52:19< travis-ci> wesnoth/wesnoth#11063 (master - 713711a : Celtic Minstrel): The build is still failing.
20160918 21:52:20< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160892284
20160918 21:52:21-!- travis-ci [~travis-ci@ec2-54-158-59-140.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 21:52:26< Aginor> if you can't do that, there's no point to porting dialogs, it's just a waste of time :/
20160918 21:52:36< celticminstrel> I think I could manage moving the theme (sidebars etc) to GUI2 somehow, but not sure how to deal with the map.
20160918 21:52:56< Aginor> make a completely new component
20160918 21:53:30-!- Appleman1234 [~Appleman1@KD119104045198.au-net.ne.jp] has joined #wesnoth-dev
20160918 21:54:44< Aginor> that'll also allow you to be independant from the editor
20160918 21:59:46-!- horrowind [~Icedove@2a02:810a:8380:10a8:21b:fcff:fee3:c3ff] has quit [Quit: horrowind]
20160918 22:00:15-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 276 seconds]
20160918 22:00:37< celticminstrel> I'm going to set aside the floating textbox for now.
20160918 22:03:53< celticminstrel> It seems as though it might require a GUI2 game UI as a prerequisite. :/
20160918 22:04:26< celticminstrel> Or some kind of arcane trickery that I can't figure out, but the former's probably the better route, really.
20160918 22:04:52 * Aginor nods
20160918 22:05:10< Aginor> and that's not a small task
20160918 22:06:50-!- tad_ [add94167@gateway/web/freenode/ip.173.217.65.103] has joined #wesnoth-dev
20160918 22:08:13< celticminstrel> I kinda hope it can be done by 1.14, but no-one has even gotten started on it, so I wonder if there's hope.
20160918 22:08:22< celticminstrel> Ah... I used "hope" twice. Meh.
20160918 22:09:04 * tad_ chuckles
20160918 22:09:36< tad_> We need to do a custom splash screen for -dev .. under the logo "Abandon all hope, ye who enter ..."
20160918 22:09:42< celticminstrel> Ooh! It's running! (The Mac build with the menus refactor.)
20160918 22:09:52-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 22:09:54< travis-ci> wesnoth/wesnoth#11064 (gui2_floating_textbox - 6d8e44a : Celtic Minstrel): The build is still failing.
20160918 22:09:54< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160897727
20160918 22:09:54-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 22:09:58< celticminstrel> What're the chances it'll work?
20160918 22:11:05< tad_> src/floating_textbox.cpp:41:2: error: unused parameter ‘gui’ [-Werror=unused-parameter]
20160918 22:11:21-!- edgrey [~edgrey@178.204.202.173] has quit [Quit: Konversation terminated!]
20160918 22:11:39< celticminstrel> That's okay, it's only a branch.
20160918 22:11:52< celticminstrel> Or is that from the master failure?
20160918 22:11:58< celticminstrel> Shouldn't be though...
20160918 22:12:11< tad_> MAster was fine when it completed for me a moment ago
20160918 22:14:10-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160918 22:14:42< irker634> wesnoth: Celtic Minstrel wesnoth:master 0e637361b11d / src/play_controller.cpp: Match code to comment https://github.com/wesnoth/wesnoth/commit/0e637361b11de9806f130ae9bd20e7815019433b
20160918 22:15:56-!- celmin [461b1bda@gateway/web/freenode/ip.70.27.27.218] has joined #wesnoth-dev
20160918 22:16:10 * celmin here through webchat so I can just click build links.
20160918 22:16:54< celticminstrel> Now let's see if the menus even work anymore.
20160918 22:17:35< celticminstrel> Hm, curious. Some do, some don't.
20160918 22:17:37< celticminstrel> And crash.
20160918 22:17:39< celticminstrel> Lovely.
20160918 22:17:47-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 265 seconds]
20160918 22:17:48-!- wedge010 is now known as wedge009
20160918 22:18:08< celticminstrel> Did I forget to call show_menu somewhere?
20160918 22:25:10< celticminstrel> Oh, I suppose only editor menus should be broken.
20160918 22:33:07-!- Duthlet [~Duthlet@dslb-146-060-179-135.146.060.pools.vodafone-ip.de] has quit [Quit: leaving]
20160918 22:46:32< gfgtdf> loonycyborg: there is a 'asio_umcd' branch in wesnoth. is that related to your work? can it be deleted?
20160918 22:46:53-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has joined #wesnoth-dev
20160918 22:46:54< travis-ci> wesnoth/wesnoth#11065 (master - 0e63736 : Celtic Minstrel): The build is still failing.
20160918 22:46:54< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/160903825
20160918 22:46:54-!- travis-ci [~travis-ci@ec2-50-17-23-137.compute-1.amazonaws.com] has left #wesnoth-dev []
20160918 22:47:05< loonycyborg> it was old gsoc project that didn't get merged
20160918 22:47:43< gfgtdf> celticminstrel: why did you remove the code in /window_builder.cpp ?
20160918 22:48:47< gfgtdf> loonycyborg: since you did it now we have no longer use for that code and can remove it right?
20160918 22:49:11< loonycyborg> not sure, probably yes
20160918 22:49:18< loonycyborg> though that was a full rewrite
20160918 22:49:21< loonycyborg> rather than a port
20160918 22:55:46< shadowm> gfgtdf: Why do you want to remove it?
20160918 23:18:31-!- rwr3000 [45888259@gateway/web/freenode/ip.69.136.130.89] has joined #wesnoth-dev
20160918 23:27:13< gfgtdf> shadowm: i thought with the new campaignd code it now cannot be merged anymore.
20160918 23:27:48-!- Greg-Boggs [~greg_bogg@2601:1c2:901:e170:ccf:8c91:43d2:fe86] has quit [Remote host closed the connection]
20160918 23:27:49< gfgtdf> celticminstrel: the travis buidl fail since your latest commits.
20160918 23:33:44< rwr3000> Quick question, what do you guys use for editing wesnoth code? For now I've just been using vim but am curious what others use, just getting started.
20160918 23:35:17< gfgtdf> rwr3000: im on windows so i cannot help you regarding vim alternatives.
20160918 23:35:30< gfgtdf> shadowm: i think you used linux to develop for wesnoth ?
20160918 23:37:43< rwr3000> I do have an older windows machine I can use if there is a good program for it. I mainly use macOS but also have a linux machine
20160918 23:39:24< gfgtdf> rwr3000: i think Xcode is the standard c++ ide on macos ?
20160918 23:40:20< tad_> rwr3000: vim is fine and I've used it a lot. For wesnoth, I've been trying out VS Code. But I still keep a console open to run the game, do git which VS Code doesn't, and, yes, to use vim when it's too much trouble to wait for VS Code to open and find the file.
20160918 23:42:12-!- rwr3000 [45888259@gateway/web/freenode/ip.69.136.130.89] has quit [Ping timeout: 240 seconds]
20160918 23:43:48< irker634> wesnoth: mattsc wesnoth:fix-tunnel-issues 738491fe46ba / src/pathfind/ (teleport.cpp teleport.hpp): [tunnel] tag: add new optional parameter allow_vision=yes/no https://github.com/wesnoth/wesnoth/commit/738491fe46ba776369f7736f97e0c33607319c38
20160918 23:46:05< mattsc> I think with that ^ I have done what I wanted to do for vision through tunnels.  https://github.com/wesnoth/wesnoth/pull/790
20160918 23:47:06< mattsc> There are a few more things I want to check, such as on-screen messages when the exit hex is occupied by an enemy ambusher etc., but other than that this was it.
20160918 23:47:20< mattsc> Unless somebody else finds something missing, of course.
20160918 23:47:50< celticminstrel> gfgtdf: That's explained in the comit message.
20160918 23:49:08< celticminstrel> ^commit
20160918 23:49:37< mattsc> Well, I should replace the spaces with tabs …
20160918 23:54:00< tad_> I'm a 'spacer' .. too many different editors .. too many different widths for tabs .. 2 here, 4 there, 8 someplace else. I like things to stay put!
20160918 23:54:30< mattsc> Yeah …
20160918 23:55:56< celticminstrel> gfgtdf: If you understand that code and why it's needed and thing that putting it back will fix Travis, then go right ahead.
20160918 23:55:58< celticminstrel> ^think
20160918 23:56:12< celticminstrel> (Though of course leaving in the FAIL instead of VALIDATE.)
20160918 23:56:28< celticminstrel> (Sorry, instead of ERROR_LOG.)
20160918 23:58:39< shadowm> gfgtdf: Yeah, but people don't destroy museums because better technologies and art have been created.
20160918 23:58:40< mattsc> tad_: so I don’t get the UNKNOWN aspect[version*composite_aspect] error in ThoT:S9 when I start from the beginning of the scenario.
20160918 23:58:48< mattsc> But I do get it if I load a saved game.
20160918 23:59:17-!- fabi [~fabi@176.7.128.129] has joined #wesnoth-dev
--- Log closed Mon Sep 19 00:00:17 2016