--- Log opened Wed Sep 06 00:00:42 2017
20170906 00:01:38< irker633> wesnoth: Charles Dang wesnoth:master d864a89a15ee / src/game_initialization/flg_manager.cpp: FLG Manager: keep all random options sorted at the top of the faction list https://github.com/wesnoth/wesnoth/commit/d864a89a15ee0159b28b5daec492d3070db31032
20170906 00:01:41< irker633> wesnoth: Charles Dang wesnoth:master d707f94830bc / data/gui/window/mp_staging.cfg: MP Staging: fixed position of gender image tooltip https://github.com/wesnoth/wesnoth/commit/d707f94830bced933a4d8e6a7d736ff541d795c5
20170906 00:01:43< vultraz_iOS> gfgtdf: ^
20170906 00:05:53-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170906 00:19:30-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170906 00:19:43-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170906 00:21:15-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170906 00:22:41-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20170906 00:28:51< irker633> wesnoth: gfgtdf wesnoth:master d08f402ef167 / src/replay.cpp: fix 'at_end()' assertion faliure when chatting in a mp game https://github.com/wesnoth/wesnoth/commit/d08f402ef1677a8389c9cdd0040d1c18a085d9f1
20170906 00:29:01< gfgtdf> Necrosporus: try that ^
20170906 00:32:43< gfgtdf> vultraz_iOS: you should change the channel topic date to 00:01 :p. some people might find it unclear which date is meant exactly.
20170906 00:35:37< celticminstrel> Yes please.
20170906 00:35:59< celticminstrel> It's technically unambiguous, mind you, but some people may still be confused.
20170906 00:36:58< celticminstrel> (ISTR reading that 00:00 is officially the beginning of the day, while 24:00 is the end of the day; IOW 24:00 Sept 16 is the same moment as 00:00 Sept 17.)
20170906 00:37:33< celticminstrel> vultraz_iOS: So, what should we do with the map generation PR if I don't get to it this week? I expect to get to it, but I might not.
20170906 00:37:37< DeFender1031> 24:00 is BS.
20170906 00:38:32< celticminstrel> https://en.wikipedia.org/wiki/12-hour_clock#Confusion_at_noon_and_midnight
20170906 00:38:46< celticminstrel> Apparently what I described is in ISO-8601.
20170906 00:39:51< DeFender1031> Yeah, as a programmer who deals heavily with date and time scheduling, I say 24:00 is BS.
20170906 00:40:06< DeFender1031> I suppose that for human communication it can be useful.
20170906 00:40:10< DeFender1031> Maybe.
20170906 00:40:21< celticminstrel> Yeah, it's probably difficult to deal with from a programming standpoint.
20170906 00:40:23-!- Bonobo [~Bonobo@61.68.158.162] has quit [Read error: Connection reset by peer]
20170906 00:40:54 * celticminstrel considers opening a thread about the Irdya calendar... where would that belong...
20170906 00:41:02< DeFender1031> Not difficult to deal with, just not meaningful, since you would just always use 00:00 from the following day.
20170906 00:41:04< celticminstrel> Ideas maybe?
20170906 00:41:48< celticminstrel> DeFender1031: I'm suggesting it would be difficult to deal with because it implies there are two ways of representing the same moment in time.
20170906 00:42:57< DeFender1031> ah, yeah, exactly
20170906 00:43:59< DeFender1031> also because 00:00-23:59 means that there are 24 possible hours (0-23) which each have 60 possible minutes (0-59). 24:00 adds a 25th possible hour, with only one possible minute.
20170906 00:45:12< DeFender1031> Basically, anything storing a date and time internally would use the 0-23:0-59 format, though on output for humans, there may be some cases where 24:00 is more useful (for example, showing the start and end time of some event starting at 11 PM and lasting an hour).
20170906 00:45:59< DeFender1031> Another reason I say that it's BS though is because at the exact moment when that minute flips over, it's no longer the previous day.
20170906 00:46:37< celticminstrel> I'd argue that that viewpoint is nonsensical.
20170906 00:46:40< DeFender1031> Meaning there is no actual instant when it's no longer 23:59 but still IS the date on which that 23:59 happened.
20170906 00:47:00< celticminstrel> There's no real boundary between one day and the next other than the arbitrary one we've defined.
20170906 00:47:01< DeFender1031> as soon as it stops being 23:59, it stops being that day.
20170906 00:47:23< DeFender1031> Well yes, certainly, but based on the rules that have been defined, what I said is true
20170906 00:47:42< DeFender1031> there isn't this schroedinger's minute which is both yesterday and today.
20170906 00:47:54< celticminstrel> A computer should worry about 24:00 only when parsing dates, I guess.
20170906 00:49:38< DeFender1031> Well, only if whatever it's parsing allows for that... And even then, it should obviously parse it such that the internal data translates into 00:00 on the following day
20170906 00:50:00< DeFender1031> And I'd be highly suspicious of anything whose output would allow for that.
20170906 00:50:10< celticminstrel> Right.
20170906 00:50:48< DeFender1031> 24:00 really IS a nonsensical time value.
20170906 00:51:28< DeFender1031> Like, imagine that we said that you were allowed to call November 1st "the 32nd of October" as an alternate name.
20170906 00:52:05< DeFender1031> I mean, what? Just because the hours and minutes are 0-indexed instead of 1-indexed that weird wrappage suddenly makes sense?
20170906 00:52:54< shadowm> celticminstrel: Writer's Forum, definitely.
20170906 00:53:05< DeFender1031> Or, better example, that's also 0-indexed. We can say that "4:60" is another valid name for "5:00".
20170906 01:02:26-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 246 seconds]
20170906 01:19:47-!- gfgtdf_ [~chatzilla@x4e32b72e.dyn.telefonica.de] has joined #wesnoth-dev
20170906 01:20:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 01:21:36-!- gfgtdf [~chatzilla@x4e368b1c.dyn.telefonica.de] has quit [Ping timeout: 248 seconds]
20170906 01:21:42-!- gfgtdf_ is now known as gfgtdf
20170906 01:24:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 248 seconds]
20170906 01:35:21-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Ping timeout: 240 seconds]
20170906 01:39:45-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170906 01:48:40-!- gfgtdf [~chatzilla@x4e32b72e.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20170906 01:51:45-!- travis-ci [~travis-ci@ec2-107-20-93-88.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 01:51:46< travis-ci> wesnoth/wesnoth#14901 (master - 1b7c36e : ln-zookeeper): The build has errored.
20170906 01:51:46< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272183014
20170906 01:51:46-!- travis-ci [~travis-ci@ec2-107-20-93-88.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 02:03:01-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20170906 02:05:35< Necrosporus> DeFender1031, on recently discussed question. My microwave owen allows to set minutes and seconds up to 99, so you can make it cook something for 1:90 which is 60+90 seconds which is same as 2:30
20170906 02:06:07< shadowm> #BadUX
20170906 02:06:34< DeFender1031> Mine too. My wife says that she purposely does 90 instead of 1:30 because she feels bad for the 9 button that it doesn't get pressed otherwise. :P
20170906 02:08:30< DeFender1031> On the other hand, since the input takes digits in sequence, in order to enter 9:00, you have to be able to enter 90 first, so if you enter 90 and then hit start, the only choices are to treat it as 90 seconds or to not allow you to start it at all at that point.
20170906 02:10:13< DeFender1031> (Side point: "microwave owen" sounds like the lamest superhero ever.)
20170906 02:13:08< vultraz_iOS> wife? :o
20170906 02:13:20< DeFender1031> Umm... yes?
20170906 02:13:47< DeFender1031> Is this surprising?
20170906 02:13:49< shadowm> vultraz_iOS is one of *those* people who believe you can't have a SO if you are on the Internet.
20170906 02:15:22< vultraz_iOS> it shouldn't be surprising (hell, dave's been married for years) yet somehow it is
20170906 02:17:34< DeFender1031> Checking my logs, I've mentioned her offhand a few times, though it seems that vult was not present any of those times.
20170906 02:18:25< DeFender1031> I'm curious though why the default assumption is unmarried.
20170906 02:19:39< vultraz_iOS> maybe I've subconsciously subscribed the the tired nerd stereotype  
20170906 02:20:42< vultraz_iOS> to*
20170906 02:20:44< DeFender1031> Perhaps, though apparently tired nerds get married too...
20170906 02:20:51< DeFender1031> :P
20170906 02:21:24< vultraz_iOS> also, it's a weird role reversal when the 19 year old kid in school (ie myself) is telling the successful married guy with a good job what to do O_O
20170906 02:22:04-!- vultraz_iOS changed the topic of #wesnoth-dev to: 1.13.9 scheduled for Sunday, September 17th, 00:01 UTC | Wesnoth Developers Channel | >>> Want to help? Go here: http://r.wesnoth.org/t42911 (and thanks!) <<< | Discord Server: https://discord.gg/tSmJS2E | Logs: http://irclogs.wesnoth.org | Bug tracker: http://bugs.wesnoth.org
20170906 02:22:14< DeFender1031> (I certanly thought it'd never happen, then all of a sudden, there was this cute girl talking about Babylon 5 and I was like "THEY DO EXIST!")
20170906 02:23:50< DeFender1031> vultraz_iOS, that's sometimes how it happens in online communities.
20170906 02:24:51< DeFender1031> Also "successful" is a relative term... :P
20170906 02:29:17< vultraz_iOS> more than I at least
20170906 02:29:19< vultraz_iOS> :P
20170906 02:29:43< DeFender1031> Dude, you're young. Give it time. Sounds like you're as successful as you ought to be at this stage.
20170906 02:30:19< vultraz_iOS> heh, yeah, well... 
20170906 02:30:57< vultraz_iOS> not exactly. but nevermind.
20170906 02:31:52< shadowm> Becoming project admin for a pre-established OSS project isn't all that great.
20170906 02:32:05< DeFender1031> One of the things that the older, more experienced software people in my community encouraged me to do when I was your age was to get involved in a popular open soure project to gain experience and some recognition. What I actually did was... a hell of a lot of gamecube. You, on the other hand, are a pretty decent programmer already and are basically managing the team for an open source project.
20170906 02:32:37< DeFender1031> And crap, I just used the phrase "when I was your age". I've become an old fart!
20170906 02:32:57< vultraz_iOS> xD
20170906 02:33:15< shadowm> You quickly find out that order is an illusion and that people just do their own thing and don't take kindly to being told that they are doing it wrong or that they are prioritizing their own interests over the project's.
20170906 02:34:16< shadowm> You also de facto assume responsibility for things that other people beyond your control do.
20170906 02:34:46< DeFender1031> shadowm, well, it's tough when people are volunteering their time. You can't dock their pay or fire them, and they are under no obligation to do what you want them to. The most you can do is reject their changes if they're bad changes, but if they're acceptable changes but just not what the project as a whole needs prioritized, there's not much recourse.
20170906 02:35:27< DeFender1031> On the other hand, a community will (or at least should) try to maintain a shared vision, common goals, etc. so as to minimize that sort of crap.
20170906 02:36:33< shadowm> There are communities that far much better in that regard (e.g. KDE). Wesnoth just doesn't have that.
20170906 02:36:36< shadowm> *fare
20170906 02:37:03< DeFender1031> Not sure I agree. KDE had a MAJOR falling out and change of vision between versions 3 and 4
20170906 02:37:22< shadowm> These days.
20170906 02:39:29< shadowm> For reference, I've been here for eleven years, and been a developer for ten, so I've gotten to experience its gradual disintegration first-hand.
20170906 02:40:00< DeFender1031> Well, if we're only talking these days, wesnoth has basically zookeeps, minstrelman, vult, gfg, jyrki, and a couple of others occasionally, with me making suggestions here and there until I finally have enough free time to actually contribute actual code, and aside from a few logistical things about code management, everyone seems to more or less have the same vision for wesnoth.
20170906 02:40:16< shadowm> And from both sides of the wall as well, since I've never stopped being a player and user of the game, unlike many of the other senior devs.
20170906 02:41:32< DeFender1031> That's all fair, but if we're comparing to apples to apples, we either have to say that both KDE and Wesnoth had their rocky periods, or that both of them have a development community with a common vision as of now, no?
20170906 02:42:03< DeFender1031> And granted, wesnoth's dev list is laughably small compared to KDE's, but still...
20170906 02:42:11< vultraz_iOS> but not a common vision of communication platforms :) :) :) :) :) :) :)
20170906 02:42:13< shadowm> Applying the common vision thing to Wesnoth, right now, is way too idealistic.
20170906 02:42:24< shadowm> vultraz no.
20170906 02:42:28< shadowm> This is not the time for that tangent.
20170906 02:42:31< DeFender1031> shadowm, what do you mean?
20170906 02:42:36< vultraz_iOS> I'm not going there
20170906 02:42:57< shadowm> I don't want to name any names, but several of the people you mentioned have wildly different visions about how to move the game forward in terms of non-code matters.
20170906 02:43:34< shadowm> And also one you didn't mention whose role in this is as irrelevant as mine right now.
20170906 02:44:37< shadowm> One of the people who you did actually mention would rather maintain the status quo and not have anything change in the slightest, even if doing so would actually help the project by attracting new people able to keep it on life support.
20170906 02:45:30< shadowm> Another of the people who you mentioned would rather make much more profound and drastic changes to the game in order to attract new players.
20170906 02:45:51< shadowm> (Partly influenced by the opinions of a couple of unseen actors who are not active developers.)
20170906 02:46:00< DeFender1031> If you're referring to who and what I think you're referring to, said person has repeatedly tried to make it clear that "no changes ever" is NOT what he thinks.
20170906 02:46:40< DeFender1031> But again, those are some of the "logistical code management" issues to which I referred.
20170906 02:47:10< shadowm> Wesnoth is not Open Source Software Management Sim.
20170906 02:47:23< shadowm> Wesnoth is a turn-based strategy game with a high fantasy theme.
20170906 02:47:38< shadowm> Ultimately, what is being built here is the latter, not the Battle for Wesnoth Project.
20170906 02:47:57< shadowm> But some tend to confuse the means with the end.
20170906 02:48:11< DeFender1031> I'm not sure I know what the former means.
20170906 02:50:49< shadowm> That ultimately, whether you're making things faster by rewriting the engine to use OpenGL 9.0 or reaching complete test coverage with over 2,000 individual unit tests is completely irrelevant if you can't ship a working product that is able to attract and entertain new people in this day and age, and keep their attention for long enough for them to consider contributing back to the project.
20170906 02:51:29< DeFender1031> true.
20170906 02:51:30< shadowm> Those people don't care if version 1.17.0 finally implements a [brew_coffee] tag, either.
20170906 02:53:07< DeFender1031> Depends. Average players may not, but content developers may, if they want to have a campaign with a lot of coffee in it and can't code the WML or Lua to brew coffee themselves...
20170906 02:53:32< DeFender1031> There's certainly a balance that's needed between making a good game and building a stable community
20170906 02:53:46< DeFender1031> I think everyone here realized that both of those goals are important
20170906 02:54:23< DeFender1031> I think people just have different ideas about how to deal with situations where a conflict between those two goals arises.
20170906 02:54:26< shadowm> I became the de facto Community Manager before Community Manager was even an official role.
20170906 02:54:34< shadowm> I assure you, I didn't have any support from the development team.
20170906 02:54:56< DeFender1031> Sigh...
20170906 02:58:35< shadowm> At the end of the day, our (historical, at least) devs only really become interested in community matters when they see an immediate benefit (adding campaigns to mainline) or feel that the privileges they've ascribed to themselves out of a sense of 'seniority' are being threatened.
20170906 02:59:41< shadowm> But otherwise it's always been a free-for-all and I found that out the hard way around the same time I became release manager.
20170906 03:13:37-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 03:18:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20170906 03:30:14< celticminstrel> I don't see vultraz_iOS answering my question?
20170906 03:30:27< vultraz_iOS> What?
20170906 03:31:12< Necrosporus> >  <gfgtdf> Necrosporus: try that ^ // compiling it. Halfway done
20170906 03:31:47< Necrosporus> Now I either need to persuade somebody to compile it too
20170906 03:32:18< Necrosporus> or find somebody who has latest master compiled already
20170906 03:34:48< celticminstrel> Also, please don't call me minstrelman... :|
20170906 03:34:51< celticminstrel> [Sep 05@8:37:33pm] celticminstrel: vultraz_iOS: So, what should we do with the map generation PR if I don't get to it this week? I expect to get to it, but I might not.
20170906 03:35:15< vultraz_iOS> If you don't get to it postpone it
20170906 03:35:26< celticminstrel> To when? 1.13.10? 1.14.0?
20170906 03:35:45< DeFender1031> celticminstrel, sorry, was just trying to avoid pinging you. Didn't realize it wouldn't be appreciated.
20170906 03:36:21< celticminstrel> It's admittedly hard to avoid pinging me, given that I'm pinged by several common abbreviations of my name.
20170906 03:38:31< DeFender1031> I think I've said stuff like "welsh poet" in the past as well... Though if you'd prefer I use one of your proper names even at the expense of pinging you then I will certainly respect that for the futuer.
20170906 03:38:35< DeFender1031> future*
20170906 03:38:50< celticminstrel> Welsh poet, heh.
20170906 03:39:02< celticminstrel> Gives a slightly wrong impression, but whatever.
20170906 03:39:18< vultraz_iOS> i've used "germanic bard" :P
20170906 03:39:30< celticminstrel> ...how the heck do you get that?
20170906 03:39:38< celticminstrel> Also vultraz_iOS you didn't answer my followup.
20170906 03:39:54< celticminstrel> I have gaelicbard somewhere in my list of fallback nicks.
20170906 03:40:07< celticminstrel> But to get from celtic/gaelic to germanic seems a bit of a stretch.
20170906 03:40:44< vultraz_iOS> yeah, I guess the celts aren't germanic
20170906 03:40:47< vultraz_iOS> celticminstrel: 1.13.10
20170906 03:41:07-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170906 03:41:21< celticminstrel> 'kay
20170906 03:41:54< celticminstrel> I don't think I mind being pinged every so often. At least I don't get an audible notification (most of the time).
20170906 03:42:08< DeFender1031> Or "boston basketball bard".
20170906 03:42:16< celticminstrel> So I guess, do whatever?
20170906 03:42:29< vultraz_iOS> heh, I have indeed also used Celtic minstrel 
20170906 03:42:35< celticminstrel> ?
20170906 03:43:00< vultraz_iOS> well, not indeed, since i guess that's different
20170906 03:43:19< vultraz_iOS> or actually no
20170906 03:43:20< celticminstrel> You've lost me.
20170906 03:43:21< vultraz_iOS> i was right
20170906 03:43:47< vultraz_iOS> celticminstrel: boston celtics 
20170906 03:44:03< celticminstrel> Ah. I don't follow any sports, so I would've never gotten that one.
20170906 03:44:31< DeFender1031> Neither do I, but I used to play NBA Jam with my brother when we were kids...
20170906 03:44:46< vultraz_iOS> Neither do I, but I know the name
20170906 03:44:48< celticminstrel> I don't play sports games, either. :P Except Wii Sports on occasion.
20170906 03:45:04< celticminstrel> Oh, unless you count racing as a sport, but even then it's not very often.
20170906 03:45:16< celticminstrel> (eg Mario Kart)
20170906 03:59:55< irker633> wesnoth: Charles Dang wesnoth:master 0882d22b2204 / src/game_initialization/flg_manager.cpp: Fixup d864a89 https://github.com/wesnoth/wesnoth/commit/0882d22b2204cc51550269e3e27139f8c2038b7d
20170906 04:03:14< celticminstrel> Ah, yeah, your comparator needs to be consistent.
20170906 04:03:43< celticminstrel> If "a < b" and "b < a" both return true for any pair (a,b), you have a problem.
20170906 04:04:25< celticminstrel> It's nice that MSVC's debug libraries actually check this for you. Other libraries are generally content to just let it go. I kinda wonder what effect it actually has though...
20170906 04:06:07< celticminstrel> (Technically this does mean you can get into trouble sorting lists of floats or doubles...)
20170906 04:06:40< JyrkiVesterinen> Well, the old comparator was returning *false* for both "a < b" and "b < a" and therefore didn't trigger Microsoft CRT's debug check.
20170906 04:07:21< JyrkiVesterinen> And for that reason, sorting lists of floating-point numbers is always safe, too.
20170906 04:08:24< celticminstrel> Oh?
20170906 04:09:09< celticminstrel> Wait, if both "a < b" and "b < a" are false, wouldn't they end up sorted as if a and b are equal?
20170906 04:10:26< JyrkiVesterinen> Indeed. That's how the CRT even knows that they're equal.
20170906 04:11:02< JyrkiVesterinen> a < b == true: a is smaller. b < a == true: b is smaller. a < b == false && b < a == false: a and b are equal.
20170906 04:11:39< celticminstrel> So basically, it didn't trigger the runtime check, but would have produced incorrect results...?
20170906 04:11:48< JyrkiVesterinen> std::sort() is unstable (i.e. may reorder equal elements), whereas std::stable_sort() is stable and will leave equal elements to their original order.
20170906 04:12:36< JyrkiVesterinen> Yes, Vultraz's old code ordered the random factions depending on whether they ended up as the lhs or the rhs parameter.
20170906 04:12:36< celticminstrel> I assume GUI2 listboxes do use the latter... (random tangent)
20170906 04:13:16< vultraz_iOS> yes, the generator uses stable_sort
20170906 04:44:15< Necrosporus> building new wesnoth thrunk is finished, it works, even though it's one commit behind now
20170906 04:44:18-!- travis-ci [~travis-ci@ec2-54-167-201-65.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 04:44:19< travis-ci> wesnoth/wesnoth#14907 (master - d864a89 : Charles Dang): The build has errored.
20170906 04:44:19< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272286635
20170906 04:44:19-!- travis-ci [~travis-ci@ec2-54-167-201-65.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 04:48:44-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 04:48:45< travis-ci> wesnoth/wesnoth#14906 (master - 138cefe : Jyrki Vesterinen): The build has errored.
20170906 04:48:45< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272223850
20170906 04:48:45-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 04:54:48< Necrosporus> So does anybody want to play with me using new build so we can see if mp campaign will work?
20170906 04:55:02< vultraz_iOS> can't right now
20170906 04:55:04< Necrosporus> It didn't work yesterday
20170906 04:58:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 05:02:55-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 248 seconds]
20170906 05:32:56-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170906 05:37:03< Necrosporus> By the way don't you think that Orc in 1.13 tutorial might be too hard for beginners? It got more gold than it was in 1.12?
20170906 05:37:25< Necrosporus> I'm not sure if it's the case though, we need some beginners to test it first
20170906 05:40:59-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 05:41:00< travis-ci> wesnoth/wesnoth#14909 (master - 0882d22 : Charles Dang): The build has errored.
20170906 05:41:00< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272328746
20170906 05:41:00-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 05:45:50-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20170906 05:51:20-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 05:51:21< travis-ci> wesnoth/wesnoth#14908 (master - d08f402 : gfgtdf): The build has errored.
20170906 05:51:21< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272292578
20170906 05:51:21-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 05:51:49< Necrosporus> And where is gfgtdf?
20170906 06:15:03< Necrosporus> What is high-fantasy btw?
20170906 06:19:46-!- JyrkiVesterinen [~JyrkiVest@195-192-251-124.s1networks.fi] has joined #wesnoth-dev
20170906 06:24:11< JyrkiVesterinen> Necrosporus: the 1.13 tutorial is doable for beginners. I can attest to that. I managed to beat it regardless of never having played Wesnoth before.
20170906 06:24:48-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170906 06:24:59< JyrkiVesterinen> And regarding high fantasy, see https://en.wikipedia.org/wiki/High_fantasy
20170906 06:24:59< Necrosporus> JyrkiVesterinen, from first time?
20170906 06:25:20< Necrosporus> Do you want to try newest wesnoth master in multiplayer?
20170906 06:25:20< JyrkiVesterinen> It took multiple attempts if I recall correctly.
20170906 06:25:27< Necrosporus> or one commit before
20170906 06:25:46< JyrkiVesterinen> Not a good time to play, sorry. I'm at work.
20170906 06:28:15-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 06:28:16< travis-ci> wesnoth/wesnoth#14903 (master - d50a7b8 : Severin Glöckner): The build has errored.
20170906 06:28:16< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272212272
20170906 06:28:16-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 07:16:02-!- irker633 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170906 07:37:22-!- atarocch_ [~atarocch@93.56.164.28] has quit [Remote host closed the connection]
20170906 07:40:24-!- atarocch [~atarocch@93.56.164.28] has joined #wesnoth-dev
20170906 07:45:06-!- atarocch [~atarocch@93.56.164.28] has quit [Client Quit]
20170906 07:48:16-!- atarocch [~atarocch@93.56.164.28] has joined #wesnoth-dev
20170906 08:02:09-!- irker923 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170906 08:02:09< irker923> wesnoth: Charles Dang wesnoth:master 9a678da618fe / src/ (3 files in 2 dirs): Use saved_game::clear() to reset instead of creating a new object https://github.com/wesnoth/wesnoth/commit/9a678da618fe73fb7828ac320a36eca2ae398637
20170906 08:02:10< irker923> wesnoth: Charles Dang wesnoth:master 5a05d7e84906 / src/gui/widgets/toggle_button.cpp: GUI2/Toggle Button: fixed modified events being fired twice on click https://github.com/wesnoth/wesnoth/commit/5a05d7e8490690a2c76d3d9ee1a4a2ab4779147d
20170906 08:02:11< irker923> wesnoth: Charles Dang wesnoth:master 86f026281bf1 / src/gui/dialogs/multiplayer/ (mp_create_game.cpp mp_create_game.hpp): MP Create Game: removed unnecessary mod-selected callback https://github.com/wesnoth/wesnoth/commit/86f026281bf13b9e03345726cc6d74b627b3d7b2
20170906 08:25:37< Necrosporus> Wesnoth freezes when I allow it to open a link in browser until the browser is closed
20170906 08:25:58< Necrosporus> specifically replay link from mp
20170906 08:35:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 08:37:04< vultraz_iOS> well thats not good
20170906 08:39:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 246 seconds]
20170906 08:40:40< Necrosporus> I guess it is not necessarily wesnoth problem though
20170906 08:41:02< Necrosporus> I mean other programs tend to freeze when I click a link and browser not started yet
20170906 08:41:45< Necrosporus> Perhaps wesnoth process forks a browser and waits for fork to finish before redrawing interface again
20170906 08:42:10< Necrosporus> Music still plays but interface doesn't redraw
20170906 09:29:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 09:31:17-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 09:31:18< travis-ci> wesnoth/wesnoth#14911 (master - 86f0262 : Charles Dang): The build has errored.
20170906 09:31:18< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272380415
20170906 09:31:18-!- travis-ci [~travis-ci@ec2-54-162-80-169.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 09:33:49-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 246 seconds]
20170906 09:38:20< Soliton> yes, wesnoth waits for the process it forked to terminate.
20170906 09:43:42< JyrkiVesterinen> Looking at the code suggests that it's platform-specific.
20170906 09:44:11< JyrkiVesterinen> On Unix-like platforms such as GNU/Linux and macOS, it waits. On Windows, it uses ShellExecute() which doesn't wait.
20170906 09:50:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 09:57:33< vultraz_iOS> JyrkiVesterinen: is there a non-waiting GNU/macOS function?
20170906 09:58:31-!- esr [~esr@wesnoth/developer/esr] has quit [Quit: WeeChat 1.4]
20170906 10:01:53< Soliton> we're explicitely waiting for the forked process.
20170906 10:02:14< Soliton> which is what we should do. thouch we don't have to do it in a blocking fashion.
20170906 10:03:19< JyrkiVesterinen> Some searching shows that we can avoid waiting for the child if we explicitly set the process to ignore SIGCHLD.
20170906 10:03:21< JyrkiVesterinen> https://linux.die.net/man/2/wait
20170906 10:03:31< JyrkiVesterinen> "POSIX.1-2001 specifies that if the disposition of SIGCHLD is set to SIG_IGN or the SA_NOCLDWAIT flag is set for SIGCHLD (see sigaction(2)), then children that terminate do not become zombies"
20170906 10:04:06< JyrkiVesterinen> (And we'd also need to remove the call to waitpid() because it becomes invalid in that case.)
20170906 10:04:34< Soliton> neat, that sounds like a good idea then.
20170906 10:05:17< vultraz_iOS> anyone on Linux who can implement/test this?
20170906 10:05:54< JyrkiVesterinen> I think I'm the most active developer who uses GNU/Linux.
20170906 10:06:16< vultraz_iOS> oh, you use that too?
20170906 10:06:38< JyrkiVesterinen> Yes. Dual booting.
20170906 10:07:15< JyrkiVesterinen> It's what I use when I test GCC, Address Sanitizer, or other similar non-MS stuff.
20170906 10:08:41< vultraz_iOS> well, that's good.
20170906 10:08:48< vultraz_iOS> I thought we had only windows devs now
20170906 10:20:14-!- esr [~esr@wesnoth/developer/esr] has joined #wesnoth-dev
20170906 10:24:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 10:28:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170906 10:38:09-!- zookeeper_ [~lmsnie@95.175.104.74] has joined #wesnoth-dev
20170906 10:38:27-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Disconnected by services]
20170906 10:38:29-!- zookeeper_ is now known as zookeeper
20170906 10:38:31-!- zookeeper [~lmsnie@95.175.104.74] has quit [Changing host]
20170906 10:38:31-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170906 10:43:10-!- gfgtdf [~chatzilla@x4e32b72e.dyn.telefonica.de] has joined #wesnoth-dev
20170906 10:51:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20170906 10:51:45-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 11:02:05< Necrosporus> gfgtdf, what is the problem with chat and [delay]? I guess it would be OK that chat messages typed during [delay] will only be sent after it ends if it's necessary for replay consistency
20170906 11:02:26< Necrosporus> [delay] is not supposed to be longer than few seconds, right?
20170906 11:02:44-!- irker923 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170906 11:06:17< gfgtdf> Necrosporus: this does not only apply to [delay] but all other animation stuff aswell, and they can indeed take a while, i have also seen verly slow wml, code that uses [delay] regulareley so that the application foesn't freere while the wml code is calculating.
20170906 11:07:55< Necrosporus> gfgtdf, and how to reproduce it?
20170906 11:08:19< gfgtdf> Necrosporus: put a [delay] duration=20000 in a start event
20170906 11:10:03< Necrosporus> I'm not sure players are supposed to chat during cutscenes... pressing keys to advance dialogue would conflict with pressing keys to chat, would it not?
20170906 11:11:43< Necrosporus> By the way, there's another thing, if wesnoth has crashed, option changes won't save
20170906 11:11:49< Necrosporus> not sure if it's bug or feature
20170906 11:12:14< Necrosporus> Like I set options to display time in chat and it resetted after I played with you and triggered a crash
20170906 11:13:04< JyrkiVesterinen> That's known. Happens very often to us developers, as killing Wesnoth in a debugger rolls back preference changes as well.
20170906 11:15:14< Necrosporus> Just got a segfault while playing with preferences
20170906 11:15:58< JyrkiVesterinen> Do you have a backtrace (or know how to obtain it)?
20170906 11:16:37< Necrosporus> I got a way to reproduce it
20170906 11:16:52< Necrosporus> I am trying to enable sound notifications for player joins and so on
20170906 11:17:27< Necrosporus> segafault occurs immediately after I click the empty square wanting to check it
20170906 11:18:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 11:19:34< Necrosporus> Actually trying to trigger a checkbox in Alerts dialogue makes a segfault
20170906 11:19:44< JyrkiVesterinen> I can reproduce it with a Windows release build I have here.
20170906 11:20:08< JyrkiVesterinen> Can't obtain a reliable stacktrace, as it's a release build.
20170906 11:20:26< JyrkiVesterinen> Can you file a bug report in GitHub?
20170906 11:21:25< Necrosporus> I do not have any accounts here
20170906 11:21:36< Necrosporus> I have release build too
20170906 11:21:39< JyrkiVesterinen> OK. I'll file the bug myself, then.
20170906 11:21:57< Necrosporus> I can build a debug build and try again
20170906 11:22:53< JyrkiVesterinen> It would be appreciated if you made a debug build, then ran it under GDB (with "gdb ./wesnoth-debug", enter "run" when it gives a prompt).
20170906 11:23:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 248 seconds]
20170906 11:23:38< JyrkiVesterinen> And when it crashes (under GDB it freezes instead of disappearing immediately), enter "bt" in GDB and uploade the output to some pastebin.
20170906 11:23:52< JyrkiVesterinen> (And then enter "quit" to exit.)
20170906 11:24:28< Necrosporus> Not bt full?
20170906 11:24:52< Necrosporus> I know how to make a backtrace
20170906 11:25:04< JyrkiVesterinen> bt full would be even better, yes.
20170906 11:25:20< JyrkiVesterinen> I'm used to short backtraces, it's just the shorter command. :P
20170906 11:28:30< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/issues/1945
20170906 11:35:17-!- stikonas_ is now known as stikonas
20170906 11:35:54< Necrosporus> JyrkiVesterinen, but it will take two hours to build
20170906 11:37:27< JyrkiVesterinen> Oh, right. You were the one with not enough memory for parallel compiling. :(
20170906 11:38:01< Necrosporus> JyrkiVesterinen, actually I have found out that I can do parallel build if I kill the browser
20170906 11:38:14< Necrosporus> But it still takes more than a hour
20170906 11:38:37< JyrkiVesterinen> Well, I guess we can just wait that another developer starts to investigate the bug, it sounds like easy to fix.
20170906 11:39:05< JyrkiVesterinen> (Myself I'll be working on not freezing Wesnoth when you open the link, as the most active GNU/Linux developer.)
20170906 11:39:06< Necrosporus> I have started the build already, it's at 20% done
20170906 11:39:32< Necrosporus> JyrkiVesterinen, by the way I guess it won't freeze if you had browser open
20170906 11:39:45< Necrosporus> Then the child will transfer URL to main browser and close
20170906 11:40:18< JyrkiVesterinen> That's what I thought as well. Which would explain why the forums/IRC/Discord haven't completely exploded because of it.
20170906 11:41:10-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Killed (wolfe.freenode.net (Nickname regained by services))]
20170906 11:41:11-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20170906 12:13:35< Necrosporus> https://forums.wesnoth.org/viewforum.php?f=56 where is review threads for 1.12 and 1.13?
20170906 12:20:04< Necrosporus> It seems memory requirements for building debug versions are above those of regular version
20170906 12:22:04< JyrkiVesterinen> It shouldn't.
20170906 12:22:09< JyrkiVesterinen> "-O1 Optimize. Optimizing compilation takes somewhat more time, and a lot more memory for a large function."
20170906 12:22:43< Necrosporus> I dunno, but build has errored three times already
20170906 12:22:58< Necrosporus> And frozen my computer even with single thread when browser was open
20170906 12:23:00< JyrkiVesterinen> It might be that code size savings from optimizations offset the memory cost of optimizations. In that case, -Os would likely be optimal for memory usage.
20170906 12:23:31< Necrosporus> And where do I put that in cmake?
20170906 12:23:44< JyrkiVesterinen> And -Og might work for debug build generation.
20170906 12:23:51< Necrosporus> -DCMAKE_BUILD_TYPE=Debug ,that I use
20170906 12:24:10< Necrosporus> But where do I put -Os ?
20170906 12:24:35< Necrosporus> Anyway, I guess its easier to wait some hours and let it finish the job
20170906 12:24:44< JyrkiVesterinen> Add -DCXX_FLAGS_USER=-Os
20170906 12:25:09< Necrosporus> But then make will restart from scratch
20170906 12:25:18< Necrosporus> If I rerun cmake
20170906 12:25:19< JyrkiVesterinen> Indeed. :(
20170906 12:25:32< Necrosporus> So I may be better off letting it finish the job
20170906 12:25:36< JyrkiVesterinen> (I'm glad to have 32 gigabytes of memory in my PC...)
20170906 12:25:39< Necrosporus> it's 25% now
20170906 12:25:56< Necrosporus> I have 3 and it was really a lot just recently
20170906 12:26:23< JyrkiVesterinen> Yeah. Hardware keeps moving forward.
20170906 12:26:55< JyrkiVesterinen> Eight gigs is still considered adequate for gaming... but there are *phones* with that much.
20170906 12:28:16< Necrosporus> Does Wesnoth count as "gaming"/
20170906 12:28:23< JyrkiVesterinen> Yep.
20170906 12:28:41< JyrkiVesterinen> But I was referring more to latest AAA games such as Call of Duty.
20170906 12:29:04< Necrosporus> If CoD is AAA, then is Wesnoth AAAAAAA ?
20170906 12:29:25< Necrosporus> Or wait
20170906 12:29:31< Necrosporus> AAA is a type of batteries
20170906 12:29:45< Necrosporus> Then if a game is bigger than AAA it will be AA?
20170906 12:30:10< Necrosporus> And if it's even bigger it would be 18650
20170906 12:50:32-!- sevu [~Shiki@p54856C20.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170906 12:56:26-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20170906 13:07:07-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20170906 13:07:14< AI0867> shadowm: did I forget about that one when setting up letsencrypt?
20170906 13:09:07< AI0867> shadowm: actually, https://status.wesnoth.org/ seems to work fine
20170906 13:11:49< AI0867> 02:44 < DeFender1031> also because 00:00-23:59 means that there are 24 possible hours (0-23) which each have 60 possible minutes (0-59). 24:00 adds a 25th possible hour,  with only one possible minute. ← leap seconds!
20170906 13:18:17< Necrosporus> I guess leap seconds are 23:60, no?
20170906 13:18:37< Necrosporus> 23:59:60
20170906 13:18:39< Necrosporus> I mean
20170906 13:19:26< AI0867> yes, but invalidates his point just the same
20170906 13:22:25< DeFender1031> no it doesn't. In practice, leap seconds are recording by actually adjusting the clock, not by adding seconds.
20170906 13:22:51< DeFender1031> recorded*
20170906 13:23:33< JyrkiVesterinen> So, do you mean that the second 23:59:59 occurs twice, then?
20170906 13:24:24< DeFender1031> either occurs twice, lasts for two seconds instead of one, or, the way places like google and others do it, EVERY second of that day is slightly longer.
20170906 13:24:50< JyrkiVesterinen> I see. Interesting.
20170906 13:25:18< DeFender1031> (there are also skipped seconds as well sometimes, in which case the inverse is true)
20170906 13:26:47< DeFender1031> I can continue this later, have to run out now for a couple of hours.
20170906 13:35:20-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170906 14:07:31-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!]
20170906 14:11:45-!- JyrkiVesterinen [~JyrkiVest@195-192-251-124.s1networks.fi] has quit [Quit: .]
20170906 14:31:20-!- irker094 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170906 14:31:20< irker094> wesnoth: Severin Glöckner wesnoth:master 0cf8c669d310 / data/core/images/portraits/elves/sorceress.png: removed some pixels in the bottom left, near the leaf of the stab https://github.com/wesnoth/wesnoth/commit/0cf8c669d310a2d5134b48d81a90dd5f58fa0b50
20170906 14:31:22< irker094> wesnoth: Severin Glöckner wesnoth:master 970d7bd72eff / data/core/units/elves/ (Druid.cfg Enchantress.cfg Shaman.cfg Sorceress.cfg Sylph.cfg): Adjusted croppings for help pictures in  the shaman line: https://github.com/wesnoth/wesnoth/commit/970d7bd72eff9583aaadb1c4670771b3e12a9d64
20170906 14:36:23< zookeeper> i think you meant "staff" :p
20170906 14:44:19-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20170906 14:50:54-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170906 15:00:21< sevu> Yes, yes ... how do I get rid of my German
20170906 15:03:41< zookeeper> i've heard it might be a lifelong affliction
20170906 15:05:31-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev
20170906 15:05:59-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20170906 15:07:21< sevu> That may be true
20170906 15:11:27-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 15:14:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 252 seconds]
20170906 15:15:10-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170906 15:18:48-!- Kwandulin [~Kwandulin@p200300E453CC3BAEA9F44CBB29BC674D.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170906 15:34:43< shadowm> AI0867: No, it's more like you never told me that you set that up.
20170906 15:35:08< shadowm> In that case I'll transition all status.w.o links to https.
20170906 15:37:31< shadowm> AI0867: I didn't know that shadowm.ai0867.net had https either.
20170906 15:38:53-!- gfgtdf [~chatzilla@x4e32b72e.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 55.0.3/20170824053622]]
20170906 15:48:34< shadowm> AI0867: It seems you don't use HSTS though.
20170906 16:01:17< irker094> wesnoth: Ignacio R. Morelle wesmere:master 6a1d7886ebff / static/docroot/index.php static/template_inside.xsl wesmere/mw/WesmereTemplate.php: status.w.o supports secure connections https://github.com/wesnoth/wesmere/commit/6a1d7886ebff570d9885040b9ee31b7075f1c6c3
20170906 16:01:28< irker094> wesnoth: Ignacio R. Morelle wesmere:develop 6a1d7886ebff / static/docroot/index.php static/template_inside.xsl wesmere/mw/WesmereTemplate.php: status.w.o supports secure connections https://github.com/wesnoth/wesmere/commit/6a1d7886ebff570d9885040b9ee31b7075f1c6c3
20170906 16:01:30< irker094> wesnoth: Ignacio R. Morelle wesmere:develop ad381d624bb2 / static/docroot/index.php static/template_inside.xsl wesmere/mw/WesmereTemplate.php: Merge branch 'master' into develop https://github.com/wesnoth/wesmere/commit/ad381d624bb2f6ed29a8cdb4ad4e18111ea466b9
20170906 16:04:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 16:04:58< irker094> wesnoth: Ignacio R. Morelle wesnoth:master 3b1083f4e787 / data/tools/ (addon_manager/html.py helptrailer.html unit_tree/html_output.py): wesmere: status.w.o supports secure connections https://github.com/wesnoth/wesnoth/commit/3b1083f4e787b136527615a328715486412f9b69
20170906 16:08:51< irker094> shikadilord: Ignacio R. Morelle valen:master cc9e24201aae / index.php: status.w.o supports secure connections https://github.com/shikadilord/valen/commit/cc9e24201aae86a642d70cd829316364e8bd6424
20170906 16:09:15-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 252 seconds]
20170906 16:14:48-!- mjs-de [~mjs-de@x4db50570.dyn.telefonica.de] has joined #wesnoth-dev
20170906 16:20:27< sevu> Are there any reasons to not compile with openmp?
20170906 16:23:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 16:42:37< Necrosporus> Final ld in wesnoth-debug takes 1.6G ram alone
20170906 16:48:27< sevu> do you mean that in relation to my question above?
20170906 16:51:11-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection]
20170906 16:53:32< irker094> wesnoth: Jyrki Vesterinen wesnoth:master b95c9996bf43 / src/ (desktop/open.cpp wesnoth.cpp): Fix freeze when launching web browser on Linux/macOS https://github.com/wesnoth/wesnoth/commit/b95c9996bf435daa4d7e64fd1ef41425ac182b95
20170906 16:56:02< Necrosporus> Nope
20170906 16:56:37< Necrosporus> Its just that I cannot build battle for wesnoth because linking stage takes 2.7G of RAM and I have 3 total
20170906 16:56:46< Necrosporus> in debug mode I mean
20170906 16:57:03< Necrosporus> I mean I probably can I make a swap file big enough and wait for a while
20170906 16:57:20-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20170906 16:58:14< sevu> A year ago I compiled on an raspberry pi 3. but I think without debugging symbols.
20170906 16:58:31< Necrosporus> JyrkiVesterinen, linking wesnoth executable in build mode makes ld process take at least 2.5 GiB of RAM
20170906 16:58:41< Necrosporus> * in debug
20170906 16:59:00< Necrosporus> It didn't finish because I used alt-sysrq-f to unfreeze my machine
20170906 16:59:17< Necrosporus> or if I didn't it would be killed by system OOM anyway
20170906 17:02:38< JyrkiVesterinen> One option might be invoking ld with --no-keep-memory.
20170906 17:02:39< JyrkiVesterinen> https://linux.die.net/man/1/ld
20170906 17:02:43< Necrosporus> OK, created 1G swap and trying again
20170906 17:02:46< JyrkiVesterinen> "ld normally optimizes for speed over memory usage by caching the symbol tables of input files in memory. This option tells ld to instead optimize for memory usage, by rereading the symbol tables as necessary. This may be required if ld runs out of memory space while linking a large executable."
20170906 17:02:46< APic> lol
20170906 17:02:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170906 17:03:25< Necrosporus> JyrkiVesterinen, if 1G swap won't be enough I will try it
20170906 17:06:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170906 17:07:22-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 17:15:35< Necrosporus> real 6m59.397s user 1m8.446s sys 0m13.327s
20170906 17:16:43< Necrosporus> Final executable takes 354M
20170906 17:16:56< Necrosporus> So it's almost as big as whole data directory
20170906 17:17:14< APic> Nice.
20170906 17:20:02-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170906 17:23:01-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 17:44:52< JyrkiVesterinen> I reproduced #1945 and added call stack to the bug report:
20170906 17:44:53< JyrkiVesterinen> https://github.com/wesnoth/wesnoth/issues/1945#issuecomment-327560321
20170906 17:57:15< irker094> wesnoth: Severin Glöckner wesnoth:master b909a8ea5609 / doc/man/wesnothd.6: update manpage: default fifodir is /var/run/wesnothd https://github.com/wesnoth/wesnoth/commit/b909a8ea5609c86679032a43db12690edfe08565
20170906 17:59:23-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170906 17:59:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 18:02:45< Necrosporus> JyrkiVesterinen, so do you have an idea why it happens?
20170906 18:03:01< JyrkiVesterinen> Not yet.
20170906 18:03:21< JyrkiVesterinen> And I'm not interested to spend more effort developing Wesnoth today.
20170906 18:03:31< JyrkiVesterinen> Other developers can look into it if they want.
20170906 18:05:48-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170906 18:06:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170906 18:09:38< Necrosporus> So the thing is that check that bool has two states failed
20170906 18:10:18< Necrosporus> line 76 of src/gui/dialogs/multiplayer/mp_alerts_options.cpp says connect_signal_mouse_left_click(*b, std::bind([&, id]() { preferences::set(id, b->get_value_bool()); }));
20170906 18:13:07< Necrosporus> And in get_value_bool assertion failed
20170906 18:13:54< Necrosporus> But why does this assertion failure result in segfault?
20170906 18:14:09< vultraz_iOS> blahh
20170906 18:14:44< vultraz_iOS> that doesn't happen for me :(
20170906 18:14:49< vultraz_iOS> (no assertion)
20170906 18:16:15< JyrkiVesterinen> That's strange. Both debug and release builds crash for me, and I'm on Windows.
20170906 18:20:20< Necrosporus> For me too, but my build is not latest now, there were several commits while I built
20170906 18:21:10< JyrkiVesterinen> None of the commits in between have anything to do with that dialog or the GUI framework.
20170906 18:33:48< Necrosporus> vultraz_iOS, so how would you track down something like that if it happened for you?
20170906 18:34:24< vultraz_iOS> try to get a stacktrace maybe.
20170906 18:34:33-!- gfgtdf [~chatzilla@x4e32b72e.dyn.telefonica.de] has joined #wesnoth-dev
20170906 18:34:36< Necrosporus> I know very little of C++ (actually almost nothing, except stuff which is shared with C)
20170906 18:35:18< gfgtdf> Necrosporus: in that case you probably cannot fix it
20170906 18:37:09< gfgtdf> JyrkiVesterinen: posed on tzhat report
20170906 18:37:23< JyrkiVesterinen> Answered.
20170906 18:38:02< vultraz_iOS> JyrkiVesterinen: oh wait, you need to check one of the boxes. Now it crashes for me
20170906 18:38:26< JyrkiVesterinen> ...were the steps to reproduce that hard to follow?
20170906 18:38:45< vultraz_iOS> I didn't read them
20170906 18:38:51< vultraz_iOS> I just read them
20170906 18:38:57< vultraz_iOS> Now I have followed them
20170906 18:40:18< gfgtdf> vultraz_iOS: assigned that report to you.
20170906 18:40:26< gfgtdf> that issue*
20170906 18:40:36< vultraz_iOS> i'm really not sure what to do about it :/
20170906 18:41:13< gfgtdf> vultraz_iOS: change [&, id] to [=, id] or just [=]
20170906 18:41:22< vultraz_iOS> oh?
20170906 18:41:29< Necrosporus> vultraz_iOS, https://pastebin.com/gWv2ZFvf
20170906 18:41:51< JyrkiVesterinen> gfgtdf: [b, id] would be my suggestion.
20170906 18:42:08< JyrkiVesterinen> I think explicitly specifying every captured variable is a good practice.
20170906 18:42:14< gfgtdf> vultraz_iOS: you cannot bing a local variable (b) by reference if the labda lives longer than the function runs.
20170906 18:42:20< JyrkiVesterinen> (For example, the teams at Rovio where I was followed it.)
20170906 18:42:21< vultraz_iOS> ah, right
20170906 18:43:27-!- atarocch [~atarocch@93.56.164.28] has quit [Ping timeout: 260 seconds]
20170906 18:43:30< vultraz_iOS> hey that does fix it
20170906 18:45:26< Necrosporus> And now you are talking about something I do not understand
20170906 18:45:36< vultraz_iOS> ok i fixed it locally
20170906 18:45:40< vultraz_iOS> will push later
20170906 18:45:45< JyrkiVesterinen> Indeed, C++ specific functionality that C doesn't have.
20170906 18:45:50< Necrosporus> What is std::bind([&, id]() supposed to mean?
20170906 18:46:01< Necrosporus> especially [] part
20170906 18:46:04< gfgtdf> JyrkiVesterinen: well, in partucular for small temporary functiosn liek std::sort(... , [](){}); i always go for [&]
20170906 18:46:32< vultraz_iOS> [] is a lambda capture statement
20170906 18:47:02< JyrkiVesterinen> A lambda is, in short, a tiny function that's defined inline.
20170906 18:47:21< JyrkiVesterinen> And capturing means that the lambda gains access to some variables inside the enclosing scope.
20170906 18:47:54< JyrkiVesterinen> [&, id] means: capture id by copy, and every other variable used within the lambda by reference.
20170906 18:48:36< JyrkiVesterinen> std::bind() is a function that creates a function object from anything callable. usually a function, but in this case a lambda.
20170906 18:49:04< vultraz_iOS> binding a lambda isn't usually done
20170906 18:51:31-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170906 18:53:50< Necrosporus> Yeah, I have used lambda in other languages
20170906 18:54:03< APic> WHAT YOU SAY !!
20170906 18:57:59-!- Kwandulin [~Kwandulin@p200300E453CC3BAEA9F44CBB29BC674D.dip0.t-ipconnect.de] has quit [Ping timeout: 246 seconds]
20170906 19:06:25< Necrosporus> So, first a function is created, which gains access to variables mentioned in its body, which are b [[ within b->get_value_bool() ]] and id, and while b is passed as pointer, id is copied, so that if  preferences::set  changes id it will still be intact... But shouldn't call to preference::set copy id into local stack anyway?
20170906 19:07:57< APic> CAPTAIN: YOU KNOW WHAT YOU DOING.
20170906 19:09:46< gfgtdf> Necrosporus: what a labda does it created an anonymous class with a () operator. the variables inside [] are the member function fo that class, and the content of {} is the content of the operator() of that class.
20170906 19:09:53< gfgtdf> lambda*
20170906 19:10:30< gfgtdf> Necrosporus: you can also use [=] or [&] in which case c++ will automatically decuce which veraibles are needed.
20170906 19:10:58< Necrosporus> Wha'ts difference between = and &?
20170906 19:12:57< gfgtdf> Necrosporus: with = the member variables of the sutrct are normal objects, and thus copied when constructing the anon object,  with [&] they are references/pointers so the objects are not copied into the anon struct.
20170906 19:13:06-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 19:13:45< Necrosporus> gfgtdf, so the segfault was caused by accessing freed memory?
20170906 19:14:23< Necrosporus> Functions which creates this lambda exits, lambda is freed but the reference remains or soething?
20170906 19:14:25< gfgtdf> Necrosporus: well freed is maybe the wrong word since it pointed to the stack, (the local variable b), but yes it's a segfault
20170906 19:15:06< Necrosporus> removed from memory
20170906 19:15:17< Necrosporus> maybe overwritten
20170906 19:15:20< Necrosporus> right?
20170906 19:15:33< gfgtdf> Necrosporus: the labda had a reference to `b`, and `b` is a local variable, it'll ccrash, just like a function retunging a referece/pointer to a local object.
20170906 19:16:32< Necrosporus> Maybe some of such errors could be found with static analyzer or something?
20170906 19:17:36< JyrkiVesterinen> A static analyzer can't easily tell that the lambda continues to be accessible after the function returns.
20170906 19:18:02< JyrkiVesterinen> Most lambdas (ones passed to std::sotr(), for example) aren't accessed after the function has returned.
20170906 19:18:04< gfgtdf> Necrosporus:  the static analyzer usually doesn't know how long the labda passed to std::bind lives, if it lives shortr than the nclosing function, passing by reference is ok.
20170906 19:18:06< JyrkiVesterinen> *sort
20170906 19:35:43< Necrosporus> Maybe poking around the game will reveal more problems then
20170906 19:38:05< Necrosporus> gfgtdf, ok, then grepping for [&, for example
20170906 19:40:18< gfgtdf> Necrosporus: you coudl but again, a [& alone is no segfault
20170906 19:40:43< Necrosporus> Yeah, but somebody who knows C++ could check if it's dangerous in the context or not
20170906 19:41:01< Necrosporus> there are only three other places with &
20170906 19:42:33-!- travis-ci [~travis-ci@ec2-54-147-171-203.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 19:42:33< Necrosporus> https://github.com/wesnoth/wesnoth/blob/master/src/gui/dialogs/preferences_dialog.cpp#L593
20170906 19:42:34< travis-ci> wesnoth/wesnoth#14912 (master - 970d7bd : Severin Glöckner): The build has errored.
20170906 19:42:34< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272510350
20170906 19:42:34-!- travis-ci [~travis-ci@ec2-54-147-171-203.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 19:44:01< gfgtdf> Necrosporus: tht oe is ok, becasue toggle_box is already a reference, altough i think it better to write &toggle_box explicitly ther
20170906 19:44:35< Necrosporus> https://github.com/wesnoth/wesnoth/blob/master/src/gui/dialogs/lobby/lobby.cpp#L775
20170906 19:45:11< gfgtdf> Necrosporus: same, but again it's better to write &window instead just &
20170906 19:45:31< Necrosporus> https://github.com/wesnoth/wesnoth/blob/master/src/gui/widgets/status_label_helper.hpp#L79
20170906 19:45:36< Necrosporus> this one is last of three
20170906 19:45:51< gfgtdf> Necrosporus: same as the one ones
20170906 19:47:32< gfgtdf> other*
20170906 19:52:05< Necrosporus> https://github.com/wesnoth/wesnoth/issues/1818
20170906 19:52:19< Necrosporus> Interesting... Can it be used for MP cheating?
20170906 19:53:09< Necrosporus> Like make your client (if you host the game) send random seed which will allow your side to get traits you want?
20170906 19:55:07< Necrosporus> or the seed from server is used when you start recruiting
20170906 19:55:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 20:02:35< Ravana_> I read that issue as units before start
20170906 20:10:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170906 20:11:11-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 20:11:29-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20170906 20:19:54-!- fabi_ [~fabi@2a02:810c:c840:2e65:95ea:6e48:d6a9:3593] has joined #wesnoth-dev
20170906 20:43:38-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170906 20:47:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 20:49:26-!- mjs-de [~mjs-de@x4db50570.dyn.telefonica.de] has quit [Remote host closed the connection]
20170906 20:58:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170906 20:58:23-!- irker094 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170906 21:03:55-!- stikonas2 [~stikonas@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170906 21:15:38-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170906 21:15:44-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170906 21:18:19-!- Oebele [~quassel@143.177.58.202] has joined #wesnoth-dev
20170906 21:21:20-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit []
20170906 21:26:18-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 21:57:09-!- Oebele [~quassel@143.177.58.202] has quit [Remote host closed the connection]
20170906 22:36:19-!- travis-ci [~travis-ci@ec2-54-224-238-221.compute-1.amazonaws.com] has joined #wesnoth-dev
20170906 22:36:20< travis-ci> wesnoth/wesnoth#14914 (master - b909a8e : Severin Glöckner): The build has errored.
20170906 22:36:20< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/272592562
20170906 22:36:20-!- travis-ci [~travis-ci@ec2-54-224-238-221.compute-1.amazonaws.com] has left #wesnoth-dev []
20170906 22:48:08< mattsc> Hey all.  I’m confused about “chance of being unscathed” in the damage calculations dialog.  I must be missing something trivial.
20170906 22:48:56< mattsc> In an attack in which neither unit can level up, what’s the difference between unscathed and the probability of the units’ current HPs?
20170906 22:49:08< mattsc> Also assuming no drain etc.
20170906 22:49:24< DeFender1031> IS there a difference?
20170906 22:49:44< mattsc> Well, my understanding is there shouldn’t be.
20170906 22:50:24< mattsc> However, if you take, e.g., a grunt vs. grunt battle and the defender can die on one hit by the attacker, the dialog displays a difference.
20170906 22:50:56< DeFender1031> I think it's just supposed to be there as a quick at-a-glance readout for that particular outcome, and the more detailed ones can be found below.
20170906 22:51:13< DeFender1031> That's... odd.
20170906 22:51:18< mattsc> So why is there a difference in the case described above?
20170906 22:51:20< mattsc> Right.
20170906 22:52:10< DeFender1031> if you want to take a screenshot so that I don't have to go set that up myself, I think seeing the dialog itself might help me puzzle through this.
20170906 22:52:21< DeFender1031> But I can think of one of two possibilities:
20170906 22:52:30< DeFender1031> 1. There's a subtle difference in certain cases
20170906 22:52:38< DeFender1031> 2. There's a subtle bug in certain cases
20170906 22:52:49< mattsc> Indeed.
20170906 22:54:11< mattsc> Screenshot: http://imgur.com/a/ECl9H
20170906 22:55:45< mattsc> Same in 1.12, btw, that’s where I found it first doing some AI work.
20170906 22:55:53< DeFender1031> Yeah, I'm putting my money on possibility #2.
20170906 22:56:33< mattsc> I’m just really surprised that something that obvious has not been pointed out before.  So that’s why I am not sure that my interpretation is right.
20170906 22:57:09< DeFender1031> Hmm... By what math could something even arrive at that value for unscathed? There's a difference of 3.5% there, which doesn't obviously correlate to any of the other values involved... hmm...
20170906 22:57:58< DeFender1031> well, that aside, if it really IS supposed to be the chance of not getting hit at all, it's odd that the calculations are run separately...
20170906 22:57:59< mattsc> AFAIK, it’s calculated by simulation.  My best guess is that something is not “truncated” correctly.
20170906 22:58:17< DeFender1031> though, I suppose that if there's some kind of drain effect involved, then the caluclations WOULD be different.
20170906 22:59:02< DeFender1031> (Though in such a case "unscathed" is a pretty useless measurement anyway, unless you really care more about not getting hit than you do about having equal or greater health by the end than you did at the start, but I digress...)
20170906 23:00:53< DeFender1031> mattsc, I'm not sure that's it though. True that if the "unscathed" thing is doing a straight "the other guy attacks this many times with this chance to hit" without accounting for "there's a possibility that the other guy will attack fewer times because he might die" then that could cause such a discrepancy, but the resulting number ought to be MUCH lower in that case, no?
20170906 23:01:45< DeFender1031> Like, if that's what was going on, they'd both end up with 16%.
20170906 23:04:37< mattsc> That’s correct.  And if you set up that case, that’s what you get.
20170906 23:05:15< mattsc> In fact, if the defender can only die with two strikes hitting, it gives the same number for both lines as well.
20170906 23:05:47< DeFender1031> that's.... odd.
20170906 23:06:19< DeFender1031> I'm pretty sure that this isn't going to be solved except by looking at code, which, as usual, I don't have time to actually do at the moment.
20170906 23:06:44< mattsc> Well, the first question is whether these two lines _should_ be the same.
20170906 23:06:53< mattsc> Maybe I am missing something.
20170906 23:07:05< DeFender1031> Right.
20170906 23:07:20< mattsc> If they should, I can check out the code also.  That’s just maths (as opposed to coding), that I can handle.
20170906 23:07:39< DeFender1031> I can't think of a reason why IN THIS MATCHUP they wouldn't be the same.
20170906 23:08:17< DeFender1031> As stated above, I could see an argument for drains and potential advancement and all of that stuff to change it.
20170906 23:08:33< DeFender1031> But in this case, either the guy gets hit or he doesn't
20170906 23:08:56< mattsc> Agreed.
20170906 23:09:18< mattsc> BTW, the reason why I found this is because I am reading up on probability theory for work and I thought it would be fun (and useful, at least for some of the AI stuff I am doing) to come up with a way of calculating these values analytically.  :P
20170906 23:10:27< DeFender1031> Wait, you mean that these values are calculated by actually running every possible battle path rather than by doing straight calculations?
20170906 23:11:18< mattsc> For anything other than the case in which neither unit can die (which is trivial, mathematically), that’s my understanding.
20170906 23:12:32< DeFender1031> True, though even in other cases it ought to be theoretically possible to still calculate it mathematically rather than operationally.
20170906 23:13:37< DeFender1031> I'm pretty sure there's a theorem that says that calculations that are simple mathematical conversions can always be done in O(1).
20170906 23:19:32< DeFender1031> By "simple mathematical conversions" I mean "conversions which involve a finite number of specific known values", or however you say that in mathese.
20170906 23:21:32< mattsc> Well, once you add things like slow, drain and/or berserk, it can get pretty complicated.
20170906 23:24:15-!- irker540 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170906 23:24:15< irker540> wesnoth: Severin Glöckner wesnoth:master 9b7db0540297 / doc/man/ (wesnoth.6 wesnothd.6): miscellaneous manpage fixes https://github.com/wesnoth/wesnoth/commit/9b7db0540297c3da8997d7f81deb70e8321352e8
20170906 23:24:38< DeFender1031> true, but I think those STILL qualify as a finite number of known values... maybe.
20170906 23:25:32< mattsc> I’m sure that’s true.
20170906 23:25:48< DeFender1031> Pretty sure you can still account for that by multiplying the chance that the guy WAS ALREADY hit into the calculation somehow.
20170906 23:28:59< mattsc> Sure.  But consider, e.g. a near-dead attacker with drain.  If he drains on the first strike, he might survive the counter strike, if not, he’ll die.  So it depends not only on whether the defender hits, but also on whether the attacker himself landed a strike first.
20170906 23:29:08< mattsc> Anyways, I don’t think we’re disagreeing.
20170906 23:29:20< mattsc> I’m not saying it can’t be done, just that it’s non-trivial.
20170906 23:29:40< mattsc> Certainly more than my brain can handle at this moment.
20170906 23:29:49< DeFender1031> Oh, certainly there are a LOT of considerations involved.
20170906 23:29:57< DeFender1031> And we're not disagreeing.
20170906 23:30:07< DeFender1031> But I'm positive that it IS mathematically possible.
20170906 23:30:32< DeFender1031> It's more than my brain can handle at the moment as well.
20170906 23:31:07< mattsc> I’m sure it is.  You can do anything with multiple applications of Bayes’ theorem.  :P
20170906 23:31:25< DeFender1031> BTW, while we're on the topic, it'd be nice if whatever did damage calculations had some method by which it could be expanded from either WML or Lua, for cases where people have created custom abilities that change the outcome possibilities.
20170906 23:40:27-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170906 23:45:02-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170906 23:47:42-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20170906 23:49:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
--- Log closed Thu Sep 07 00:00:43 2017