Time Nick Message 01:16 Megaf https://github.com/minetest/minetest/issues/4553#issuecomment-332057444 05:48 nerzhul merging #6415 in ~5 mins 05:48 ShadowBot https://github.com/minetest/minetest/issues/6415 -- Add mute setting (toggled by the mute key and in the volume menu) by DTA7 06:21 nerzhul i just pushed #6466, maintenance PR on jsoncpp + our jsoncpp usage (which is deprecated since a long time ago in jsoncpp but was just properly reported recently 06:21 ShadowBot https://github.com/minetest/minetest/issues/6466 -- Update JsonCPP to 1.8.3 by nerzhul 06:23 basicer nerzhul: Awesome, JsonP was giving me some headachs on OS:X, ill give the new version a shot. 06:25 nerzhul basicer, 1.8.0 is working nicely on FreeBSD, i don't see any problem with it :) 06:26 basicer It complains about not finding a constructor, its a pretty old clang though. 06:26 nerzhul this problem is not due to jsoncpp 1.8 but your compiler, use clang 3.4 at least 06:27 nerzhul clang 4.0 is the best idea, or 5.0 06:29 nerzhul i push a little fix for preview mod, because remote servers can be slow and mod channels not ready yet 07:01 lhofhansl Hi... Please check out #6460 and #6447 07:01 ShadowBot https://github.com/minetest/minetest/issues/6460 -- Allow setting specific administrator name by lhofhansl 07:01 ShadowBot https://github.com/minetest/minetest/issues/6447 -- Prefer loading blocks near predicted future player position by lhofhansl 07:02 lhofhansl I tweaked #6447 a lot and made it much smarter. It now takes into account where the player actually looks to optimize what blocks are retrieve (by adjusting fov, wanted_range, and the predicted player position) 07:02 ShadowBot https://github.com/minetest/minetest/issues/6447 -- Prefer loading blocks near predicted future player position by lhofhansl 07:03 lhofhansl 6460 might cause some philosophical discussion, but it does fix the problem at hand - namely that you cannot start a local game and not be an administration, which changes game behavior. 07:18 basicer I get the error on 5.0.0, Undefined Symbol "Json::Value::append(Json::Value&&)" 07:24 rdococ hi 07:51 Hijiri basicer: did you update clang? 07:52 Hijiri and by 5.0.0 do you mean 0.5.0-dev? 08:16 basicer I got 5.0.0 from homebrew as the one apple ships doesnt support thread_local 08:17 basicer by 5.0.0 I mean clang's version 08:19 sfan5 basicer: try with -DENABLE_SYSTEM_JSONCPP=1 08:19 sfan5 (needs system-wide jsoncpp install obviously) 08:19 sfan5 ^ nerzhul yet another user has problems with jsoncpp :/ 08:19 nerzhul basicer, very strange freebsd supports thread_local 08:20 sfan5 could upgrading fix it? 08:20 nerzhul thread_local is part of C++11 standard 08:20 basicer Im on nerzhul's upgraded branch 08:20 nerzhul my branch doesn't change anything about compat with C++03 compilers :) 08:20 nerzhul what is your compiler version ? 08:20 basicer Apple be all like, we dont want to support thread_local until we can make it appley. 08:20 nerzhul and mac osx version ? 08:20 basicer OSX 10.10 08:21 basicer clang version 5.0.0 (tags/RELEASE_500/final) 08:21 nerzhul https://github.com/minetest/minetest/pull/6466 build pass on travis with macosx 10.9 08:21 nerzhul i use clang 5.0 too on my archlinux 08:21 nerzhul Apple LLVM version 8.0.0 (clang-800.0.38) 08:21 nerzhul here is the clang version used 08:22 nerzhul you should use system clang version it seems 08:22 nerzhul thread_local is supported in clang since 3.4 08:22 basicer system clang doesnt support thread_local on OSX 08:22 nerzhul then why the build works on travis ? 08:23 nerzhul sfan5, please note we have thread_local in mt core engine too, i added it on some static to make them linked with thread permitting to reload them when embedded server restarts :) 08:24 sfan5 nerzhul: it's not just thread_local http://irc.minetest.net/minetest-dev/2017-09-26#i_5088132 08:25 basicer sfan5: System JsonCPP does the trick. 08:25 nerzhul sfan5, thread_local + move constructor 08:25 sfan5 hm 08:26 sfan5 why does that cause a problem on so many systems? 08:26 nerzhul i don't know , it's a little bit random it seems 08:26 sfan5 also we should look at -DENABLE_SYSTEM_JSONCPP=1 being default 08:27 basicer My OS 10.10 has system clang of AppleClang 7.0.2.7000181 08:27 nerzhul please switch to 8.0 08:27 basicer How did you get a new XCode for 10.10? 08:28 nerzhul sfan5, any idea ? i'm not very macosx poweruser 08:28 sfan5 nope 08:28 sfan5 also just tested rn: -DENABLE_SYSTEM_JSONCPP=0 works on my machine 08:29 sfan5 but i've heard of it breaking with debian 08:29 nerzhul #467 clang 5.0 works fine here (linter is just not happy) 08:29 ShadowBot https://github.com/minetest/minetest/issues/467 -- Peformance is much better when not focusing on the window 08:29 nerzhul #6467 08:29 ShadowBot https://github.com/minetest/minetest/issues/6467 -- Update clang from 4.0 to 5.0 by nerzhul 08:31 nerzhul basicer, https://developer.apple.com/xcode/ 08:34 basicer It builds fine on the latest osx/xcode ofcourse. 08:34 nerzhul what is the current latest macosx version ? 10.13 no ? 08:35 nerzhul 10.10 is 3 years old 08:35 nerzhul when you tried to use clang from homebrew did you set the CMAKE_CXX_COMPILER & CMAKE_C_COMPILER variables ? 08:35 basicer Sure is, but if you build on an old machine, generally your builds are forward compatible. 08:36 basicer I put the homebrew one first in my path. 08:36 nerzhul and what did you see in cmake output on configure step ? 08:36 basicer It found the right path and all that. 08:37 nerzhul ok 08:37 basicer And it works fine with system json. 08:37 nerzhul then i think it's libstdc++ which is not compatible and XCode 8 should update it i think 08:44 sfan5 basicer: can you test if it still happens with this applied? https://github.com/minetest/minetest/pull/6466.patch 08:44 sfan5 (and -DENABLE_SYSTEM_JSONCPP=0 obviously) 08:44 nerzhul sfan5, i think this will not change anything as it's just a jsoncpp minor update 08:44 nerzhul but if it works maybe sfan5 will approve and we can merge heh 08:44 sfan5 worth a try :P 08:46 basicer You sure travis is on 10.9 ? 08:47 basicer The latest build log implies 10.11 08:49 nerzhul basicer, i think i'm wrong 08:49 nerzhul and you are right 08:49 nerzhul sorry i'm tired :p 08:49 basicer sfan5: Patch gives the same error. 08:49 sfan5 aw 08:49 nerzhul sounds logical 14:53 paramat celeron55 can you consider banning kilbith from commenting in the MT github repos, or warn him? https://github.com/minetest/minetest/pull/6447#issuecomment-332140632 https://github.com/minetest/minetest/pull/6447#issuecomment-332152047 these personal attacks have been going on for 3+ years now 14:54 rubenwardy +1 for at least a warning 15:08 celeron55 useful contributions by kilbith have also been going on for 3+ years now 15:08 celeron55 i wish there was a trivial way to handle something like this 15:09 celeron55 kilbith is of the type who doesn't care about warnings 15:12 celeron55 is there even a way to do anything like this? 15:13 celeron55 he's playing a game, and i need to add a cooldown to his action in the game 15:14 rubenwardy he may have had useful contributions, but paramat has had fairly essential contributions - second only to you across MT + MTG 15:16 celeron55 it would be a nice contribution from paramat if paramat found a way to just ignore kilbith 15:16 celeron55 github's tool for this is blocking kilbith from the entire organization 15:17 celeron55 which means he's kind of fucked too much: https://help.github.com/articles/blocking-a-user-from-your-organization/ 15:17 rubenwardy yeah, that would be a useful feature 15:17 celeron55 and there's no way to set a timeout either, outside of manually removing the blocking 15:18 celeron55 i'm not going to do that, it's too much 15:18 celeron55 i would block comments from him if it was possible - it isn't 15:19 rubenwardy hmmm 15:19 rubenwardy https://github.com/settings/blocked_users 15:20 celeron55 paramat can personally block kilbith, but it doesn't cause an ignore for kilbith in public comments 15:20 rubenwardy damn, can still see comments 15:20 celeron55 github basically doesn't have proper moderation tools 15:20 rubenwardy we could edit/delete any abusive comments 15:21 rubenwardy github has an edited flag, so is transparent 15:21 celeron55 he can edit them back any time he wants 15:21 celeron55 github's system is completely open to abuse like this 15:21 celeron55 and kilbith probably knows it 15:23 celeron55 i could start throwing a dice at every occurence like this and do the perma-everything-ban on 6 15:25 celeron55 kilbith genuinely thinks paramat is completely incompetent and expects me to throw out paramat instead of banning himself 15:26 celeron55 i.e. he doesn't say it as a personal attack really, it's his opinion and when it comes to stating opinions, he's a dick 15:27 celeron55 i don't know if this makes a difference to anyone 15:28 celeron55 not sure if it does to me 15:28 rdococ yay, celeron! 15:29 celeron55 (as a matter of fact, he's not nice to me either) 15:33 paramat yeah blocking is too much i agree, i no longer request that 15:35 celeron55 posted my solution to the thread 15:36 celeron55 https://github.com/minetest/minetest/pull/6447#issuecomment-332238618 15:38 paramat c55 that makes sense, no problem, we can just delete the comments. however he does say these things as a personal attack, he'll take any remote opportunity possible, it's just the usual years-long thing 15:39 celeron55 i'm pretty sure i've told kilbith what i think of his attitude towards you 15:39 paramat anyway i'll just ignore :) 15:42 celeron55 maybe by researching your C++ comments better you can get some respect from kilbith 15:42 celeron55 might not seem like a very fun proposition but it's a thing you can do... 15:42 paramat i don't care about his respect :) 15:43 celeron55 well, from me also 15:43 celeron55 and others 15:44 rubenwardy compsci/programming is a game of continuous self-improvement 15:44 paramat well i'm improving slowly, i don't pretend to be a competant general programmer, i have specific abilities 15:44 rubenwardy doesn't justify kilbith what-so-ever 15:45 celeron55 of course it doesn't 15:45 celeron55 the need for improvement is obvious without comments by dicks 15:46 Shara Some of those comments can be pretty offputting to contributors in general as well, no matter who they are directed at. 15:47 paramat i'm probably the odd one out here because i have had no programming education and didn't own a computer before 6 years ago 15:48 Shara paramat: you're contribution and the good you've done MT is obvious anyway. 15:48 paramat before that the last time i owned a computer was 1994 programming simple BASIC 15:48 celeron55 just learn fast enough as you go and nobody will notice 15:48 celeron55 MT was my first 3D project, literally 15:48 Shara your* 15:49 paramat thanks :) 15:49 rubenwardy I find the best way to learn is to do lots of things. ie: things separate from MT 15:50 paramat i'm also not a computer geek (not meant in a negative way) like all other devs 15:50 Krock 8) 15:50 rubenwardy you spend 4 hours a day doing Minetest, if that doesn't qualify what does? :D 15:51 paramat heh 17:41 nerzhul merging #6466 & #6462 in ~ 10 mins 17:41 ShadowBot https://github.com/minetest/minetest/issues/6466 -- Update JsonCPP to 1.8.3 by nerzhul 17:41 ShadowBot https://github.com/minetest/minetest/issues/6462 -- Update version correctly again by SmallJoker 20:40 paramat there's a Json deprecation warning during compilation in latest master "minetest/lib/jsoncpp/jsoncpp.cpp:3679:34: warning: ‘Json::Value Json::Value::removeMember(const char*)’ is deprecated" 20:42 nerzhul i missed a warning ? will look 21:14 paramat Krock i retested #6440 , still needed 21:14 ShadowBot https://github.com/minetest/minetest/issues/6440 -- CAO footstep sounds: Reduce gain to balance volume by paramat 21:16 Krock ok 21:33 paramat hmm for floats should we use std::min/max or std::fmin/fmax? 21:39 nerzhul std::fmin/fmax 21:41 paramat ok thanks, i prefer that too 21:42 paramat i'll correct one of my recent commits then 21:53 Hijiri client-side translations are not in a stable version right? 21:55 nerzhul nore, ^ 22:12 paramat #6469 trivial 22:12 ShadowBot https://github.com/minetest/minetest/issues/6469 -- Fix recent commit: std::max -> std::fmax for floats by paramat 23:23 sofar there's a bunch of dumb minetest clients that are requesting HTTP transfers from my media server incorrectly 23:24 sofar "Minetest/0.4.16 (Windows/6.1-Service_Pack_1 x86_64)" 23:24 sofar weirdly enough, it's not even android! 23:25 sofar it's requesting filename.png while everything is only available by hash.png 23:26 sofar and then a few days later that same client comes back and does a proper convo with my media server 23:26 sofar very weird 23:40 rubenwardy I found that happens if the index.mth isn't recognised correctly 23:46 sofar weird thing is that didn't even bother POST ing index.mth before it started asking for files 23:47 sofar oh well, 600 404's for them 23:49 rdococ kik 23:49 rdococ lol*