Time  Nick        Message
06:03 Zeno`       nerzhul, I just noticed. What's the go with using #pragma once?
06:03 Zeno`       I thought we didn't use non-standard stuff without a fallback
06:04 nerzhul     it's standard
06:04 Zeno`       in what standard?
06:04 nerzhul     pragma once in C++11
06:04 Zeno`       I don't think it is...
06:04 nerzhul     it is
06:05 Zeno`       https://stackoverflow.com/questions/23696115/is-pragma-once-part-of-the-c11-standard
06:05 Zeno`       "https://stackoverflow.com/questions/23696115/is-pragma-once-part-of-the-c11-standard [...]"
06:05 Zeno`       oops
06:05 nerzhul     you are right it seems commonly used by all compiler but not in standard
06:05 nerzhul     it's supported by clang, gcc and visual studio
06:06 red-NaN     to qoute wikipedia "In the C and C++ programming languages, #pragma once is a non-standard but widely supported preprocessor directive"
06:06 nerzhul     maybe we can use it ? :)
06:07 nerzhul     i didn't remember it's just used by compiler and not in the C++11 standard, maybe we can discuss this on meeting, i stop adding it to my next commits and we will choose if use it or no ?
06:08 nerzhul     if all major compiler uses it it's not standard but who compiles with other thing than GCC/Clang/VisualStudio/XCode :p
06:08 nerzhul     https://msdn.microsoft.com/en-us/library/4141z1cx.aspx
06:08 nerzhul     supported on VS 2008
06:08 nerzhul     https://msdn.microsoft.com/en-us/library/4141z1cx(v=vs.90).aspx
06:09 Zeno`       I don't know if we should use it or not. I know everything supports it but it just /feels/ wrong hehe
06:09 Calinou     pragma once never was standard, but every non-embedded compiler on Earth supports it
06:09 Calinou     GCC, MSVC, Clang, ICC
06:10 Calinou     and has been so for years
06:10 nerzhul     the pragma once has one advantage: no name collision on #ifndef #define #endif and only one tag :)
06:10 Zeno`       that's true but the way they implement it is implementation defined behaviour (by definition)
06:10 Zeno`       nerzhul, that's the big advantage yes
06:11 nerzhul     yeah, but as it seems used since a decade i hope it's stable :
06:12 nerzhul     and generally compilation is a little bit more faster on header because compiler stop reading header file on pragma once whereas #ifndef #endif requires to find the ending tag
06:12 Zeno`       I guess we just come to a decision and document that decision somewhere
06:13 Zeno`       yeah I think on gcc at least compile times are not improved (or only very slightly if they do) because of optimisations in the compiler for header guards
06:13 nerzhul     yeah, i think it's interesting to update our rules if we can use some commons things like a pragma once :) it's not a fixed document, if majority agreed let's g
06:13 Zeno`       but the name collision thing is a compelling argument /to/ use #pragma once
06:13 nerzhul     Zeno`, maybe it's why gcc is so slow to compile against clang :p
06:14 Zeno`       (just in case I wasn't clear; I'm not against using it, I'm against using non-standard extensions unless they're discussed and agreed upon)
06:14 Zeno`       nerzhul, I think that's more because gcc is the titanic and clang is a super cruiser heh
06:17 Zeno`       well I'm ok using it I guess
06:18 nerzhul     it's good to report the point :)
06:18 Zeno`       the benefits do outweigh anything unbeneficial anyway
06:18 nerzhul     yeah
06:22 nerzhul     :)
06:23 nerzhul     Zeno`, can you look at #5924 & #5933 please ?
06:23 ShadowBot   https://github.com/minetest/minetest/issues/5924 -- Have the server send the player list to the client by red-001
06:23 ShadowBot   https://github.com/minetest/minetest/issues/5933 -- [CSM] Add function to get player privileges by red-001
06:23 Zeno`       I can only have a quick look atm, not a full review
06:23 nerzhul     no problem, 5933 is relatively small, 5924 is a little bit more big but it's useful :)
06:26 Zeno`       5933 is pretty straightforward
06:26 Zeno`       no surprises in that one
06:27 Zeno`       I'm not on my main computer atm, so cannot easily browse source code
06:27 Zeno`       https://github.com/minetest/minetest/pull/5933/files#diff-7b00a65f007af40388e565e5c34d8461R334
06:27 Zeno`       that can't be const?
06:27 Zeno`       anyway, that's my only comment on 5333
06:28 nerzhul     the client pointer ? no
06:28 Zeno`       how come?
06:29 nerzhul     because we can write some things on this object if needed
06:29 nerzhul     Client *client = getClient(L);
06:29 nerzhul     client->makeScreenshot(client->getDevice());
06:29 nerzhul     for example :)
06:29 Zeno`       it's in a function though
06:30 nerzhul     do you talk about getClient or the stored pointed itself ?
06:30 Zeno`       only ->getPrivilegeList() is used
06:30 Zeno`       the pointer
06:30 nerzhul     then const Client *client = getClient(L);
06:30 nerzhul     @Zeno-
06:30 nerzhul     ?
06:30 nerzhul     oops ty gh :p
06:30 Zeno`       Client *client = getClient(L);   <--- why not   const Client *client = getClient(L);
06:31 Zeno`       unless getPrivilegeList() is not getPrivilegeList() const of course
06:31 nerzhul     it's sont :p
06:31 nerzhul     const
06:31 nerzhul     const std::unordered_set<std::string> &getPrivilegeList() const
06:31 nerzhul     i though we don't care about constness except on references and functions ? :o :)
06:32 nerzhul     then take your time heh it's better to have your right PR
06:32 nerzhul     i add #5928 to the reviews, i fixed the windows build :)
06:32 ShadowBot   https://github.com/minetest/minetest/issues/5928 -- C++11 patchset 5: use std::threads and remove old compat layer by nerzhul
06:33 Zeno`       I care though :D
06:34 nerzhul     for me no problem to add const on pointers which can be
06:34 Zeno`       :+1: on 5333 aside from my comment anyway
06:34 Zeno`       I'm driving and then will review the others
06:34 nerzhul     168 PR
06:35 nerzhul     :(
06:35 Zeno`       I can write a script to close them all if you want
06:35 nerzhul     lol
06:35 Zeno`       heh anyway bbiab :)
06:40 * red-NaN   just creates another PR anyway
06:50 red-NaN     #5940
06:50 ShadowBot   https://github.com/minetest/minetest/issues/5940 -- Use a settings object when generating world.mt and set player_backend to sqlite. by red-001
07:04 nerzhul     merging #5933 in ~10 mins (ptr fct fixed)
07:04 ShadowBot   https://github.com/minetest/minetest/issues/5933 -- [CSM] Add function to get player privileges by red-001
09:15 tenplus1    hi folks
09:15 tenplus1    can we add this now that 0.4.16 is outta the way: https://github.com/minetest/minetest_game/pull/1543
09:16 tenplus1    creative.is_enabled_for(name)  supplies player name but isnt currently used, so this pull gives it one by adding 'creative' priv to survival mode games
11:13 kilbith     https://github.com/minetest/minetest_game/pull/1543#issuecomment-306741963
11:13 kilbith     what the hell
11:13 kilbith     Zeno` is not allowed to approve
11:19 Zeno`       excuse me?
11:19 kilbith     you're not _game maintainer
11:20 Zeno`       bad luck. I approved anyway
11:20 kilbith     I think you didn't see it was MTG's PR
11:20 kilbith     well, yet another time you're breaking the rules
11:20 Zeno`       I didn't see, no. But take it to management
11:20 Zeno`       what rule?
11:20 kilbith     that only the maintainers can officially approve
11:20 kilbith     also, go do something productive instead of fouling around
11:21 Zeno`       stop trolling, please
11:21 kilbith     reminding you the rules is not trolling, sorry
11:22 Zeno`       the mtg maintainers are free to ignore the approval
11:22 Zeno`       they don't need reminding
11:22 kilbith     you're not allowed to add such label
11:23 Zeno`       technically, no.
11:23 Zeno`       But they can fix it
11:23 kilbith     we are not going to clean your shit frequently
11:23 Zeno`       you?
11:24 kilbith     here is all you have done for the engine in 6 months: https://github.com/minetest/minetest/commits/master?author=Zeno-
11:24 Zeno`       it's not your responsibility to clean things up. Sorry.
11:24 Zeno`       kilbith, that shows nothing about reviewing stuff etc etc etc
11:24 kilbith     it's my responsability to notify the maintainers
11:25 Zeno`       well, notify them!
11:25 kilbith     they are now
11:25 kilbith     Zeno`, you didn't review much from what I've seen
11:26 kilbith     you spent more time in -hub to shitposting
11:26 Zeno`       why are you attacking me?
11:26 kilbith     I only do a resume of your actions
11:27 Zeno`       If you cannot be in this channel without resorting to being an arsehole towards devs then I will ban you.
11:27 Zeno`       So, stop it. Please.
11:27 kilbith     you are the "devs"?
11:28 kilbith     *you* are being the arsehole breaking rules
11:28 Zeno`       I am "a" dev, yes
11:28 kilbith     err, sorry
11:28 red-NaN     oh give it a rest
11:28 kilbith     *who are the *devs*?
11:28 red-NaN     it's just a lable
11:28 red-NaN     label*
11:29 kilbith     someday I'll just disclose what you wrote about c55 in PM
11:29 sfan5       give it a rest
11:29 Zeno`       kilbith, and I'm sure you won't fake the logs. Get a grip.
11:30 kilbith     no I won't, I actually take screenshots
11:31 red-NaN     I mean screenshots aren't much harder to fake
11:31 Zeno`       sfan5, please remove my approval to an MTG PR if you feel it necessary. I did not notice that it was part of MTG and not engine
11:31 Fixer       we need more MTG maintainers
11:31 Fixer       MTG is so snail
11:32 kilbith     threatening of banning me when it's actually me who invited you to get back in MT...
11:32 Zeno`       There was a request to review in here with a link
11:32 Zeno`       I did so
11:32 Zeno`       end of story
11:32 kilbith     then why did you officially approve?
11:33 Zeno`       I just clicked the link that was provided in this channel (minetest-dev)
11:33 Zeno`       I reviewed it and approved
11:33 Zeno`       what don't you understand?
11:34 kilbith     I notified you that you weren't allowed to add such label but refused to take it into consideration
11:34 red-NaN     how horrifying a label got added to a PR
11:34 kilbith     that show your intend to actually break the rules, right?
11:34 red-NaN     how could something so horrible happen?
11:34 kilbith     I'm talking to Zeno, not you
11:34 red-NaN     then use PM
11:35 red-NaN     you don't get to have private conversations on a public forum
11:38 Zeno`       This is ridiculous. Being attacked like this for an honest mistake is stupid
11:39 nerzhul     wtf at lunch time, i think i should not have added a whisky with my french fries
11:39 kilbith     it wasn't a mistake since you *refused* to remove the label yourself
11:39 Zeno`       If celeron55 wants so much to separate MTG from minetest then it should be in a different organisation and repo
11:39 Zeno`       Removing the label was not a mistake
11:39 nerzhul     Zeno`, no need it just needs to create two github groups and make them have different power on each repository
11:39 Zeno`       I won't be bullied by non-devs into removing or doing anything
11:40 kilbith     in the past you wanted to allow more people to approve PRs in MTG
11:40 Zeno`       that's an entirely different issue
11:40 nerzhul     red-NaN, can you explain me the purpose of #5940 exactly ?
11:40 ShadowBot   https://github.com/minetest/minetest/issues/5940 -- Use a settings object when generating world.mt and set player_backend to sqlite. by red-001
11:40 celeron55   is kilbith a minetest_game developer?
11:41 kilbith     no
11:41 red-NaN     the idea was to cleanup the code a bit and set the player_backend so new worlds don't use the files backend
11:41 celeron55   then i don't think it's your problem
11:41 kilbith     celeron55, I'm on the #6 top-contributer on it ;)
11:41 ShadowBot   https://github.com/minetest/minetest/issues/6 -- Apples on the trees can not be eaten
11:41 kilbith     so talk to me with respect, ok?
11:41 celeron55   if minetest_game devs consider what Zeno does unwanted, then they can note it to me and i'll ask Zeno to stop (or otherwise solve it)
11:42 nerzhul     kilbith, stop to be rude PLEASE.
11:42 kilbith     what's rude is breaking the rules
11:42 red-NaN     and it would be nice to get some feedback on #5941
11:42 ShadowBot   https://github.com/minetest/minetest/issues/5941 -- [CSM] allow the client to access and modify settings in the `client.` namespace. by red-001
11:43 nerzhul     Zeno` is responsible and other dev too, they will see he isn't allowed and fix it.
11:43 celeron55   i agree Zeno probably shouldn't touch minetest_game but i've given out minetest_game management for a purpose, and the purpose is that i don't need to care 8)
11:43 nerzhul     celeron55, can you create 2 teams and split the rights ?
11:43 celeron55   i probably should
11:43 celeron55   maybe in a few weeks when i'm less busy
11:44 nerzhul     we already have website + core, it miss the mtg + core separation heh
11:44 nerzhul     wow red-NaN i didn't notice we don't use our backend if file doesn't exists, nice shot in the old code heh
11:45 Zeno`       nerzhul, that's right. And sfan5 has removed the label. It's not up to non-devs to come in here and attack me for making a mistake
11:45 Zeno`       I clicked on a link when requested for review. *shrug*
11:45 kilbith     seriously dude
11:45 nerzhul     red-NaN, you also enable player_backend to sqlite, nice, we are in 0.5 it's good
11:45 kilbith     I don't have an official role on MTG but I was a major contributor on it, so yes I care for that project
11:46 kilbith     I'm the n°6 of MTG, so fuck off
11:46 Zeno`       kilbith, then MTG review requests should probably not be allowed in this channel since it can so easily lead to honest mistakes
11:46 red-NaN     ^
11:47 red-NaN     split mt_game off into a new channel
11:47 kilbith     why not
11:48 nerzhul     generally there is no discusison about MTG here, and also it's minetest-dev not minetest-core-dev i'm neutral and don't care about this
11:48 nerzhul     rights are sufficient on GH to prevent bad actions
11:49 nerzhul     Zeno`, can you review #5928 #5940 and #5924 please ? (it's core you are allowed heh)
11:49 ShadowBot   https://github.com/minetest/minetest/issues/5928 -- C++11 patchset 5: use std::threads and remove old compat layer by nerzhul
11:49 ShadowBot   https://github.com/minetest/minetest/issues/5940 -- Use a settings object when generating world.mt and set player_backend to sqlite. by red-001
11:49 ShadowBot   https://github.com/minetest/minetest/issues/5924 -- Have the server send the player list to the client by red-001
11:49 DS-minetest red-NaN: is there yet settingtypes.txt for csm?
11:49 red-NaN     not yet
11:50 red-NaN     well it shouldn't be too hard to add
11:52 red-NaN     DS-minetest, since you are a major CSM modder it would be good to get some feedback on the API
11:52 red-NaN     for 5941
11:52 DS-minetest :D
11:53 red-NaN     it forces modders to mannually added "client." to every setting they want to use
11:54 red-NaN     I could make it do that automaticly but then it wouldn't match wants really inside the config file
11:54 nerzhul     red-NaN, it's a strange appoach but i should review it carefully
11:54 red-NaN     and what the server would see in singleplayer
11:55 DS-minetest some settings might be ok for the server to show imo
11:55 red-NaN     so what I did was basicly implement a reverse version of the "secure." namespace
11:55 red-NaN     DS-minetest, it would be but it would be hard to auto generate that list
11:56 nerzhul     oh there is a secure namespace i didn't know, what do we store inside it ?
11:56 red-NaN     so it would be best if settings that the client should be able to access a slowly moved to the "client." namespace
11:56 red-NaN     nerzhul, mod security set up
11:56 red-NaN     so lists of mods that can access insecure functions
11:56 nerzhul     okay
11:58 red-NaN     I **think** my isClient function should be safe from mod tampering but thats because I based it on the existing isSecure function
12:01 DS-minetest but clientmods that don't come from client shall never access insecure functions, right?
12:02 DS-minetest a configurable list of settings that can be accessed by clientmods could be good
12:02 red-NaN     in the current security set up only *bultin has any special privileges
12:03 red-NaN     but if this isClient function is secure we could hopefully remove those extra privileges
12:09 DS-minetest would it be simply possible to make the functions always access settings starting with `client.` if there's no special prefix (like `global.`)?
12:10 DS-minetest or `#` since that's normally not allowed
12:11 DS-minetest so, `#G` or something like that
12:11 red-NaN     and then make CSM settingtype.txt when it's added add "client." to the start of setting names right?
12:12 DS-minetest yes
12:13 red-NaN     well I could do that
12:14 red-NaN     alright I will change that
12:14 red-NaN     thanks for the input
13:19 Zeno`       For everyone's information, kilbith is being so aggressive towards me because I would not change tab_credits.lua to his liking and refused to make changes after he threatened me. It's one of the reasons I closed the PR
13:20 kilbith     ...
13:20 kilbith     I'm not going to add oil to the fire
13:21 Zeno`       well, it's true
13:37 Sokomine    latest mt does not compile for me:  minetest/src/convert_json.cpp:65: undefined reference to `Json::Value::Value(Json::Value&&)'   any ideas of what i did wrong?
13:38 Thomas-S    Sokomine: Are you using C++11? I think Minetest was updated recently to require C++11.
13:43 Sokomine    Thomas-S: seems to be gcc 4.9.2 that i'm using. not sure about C++11 support
13:44 Zeno`       make clean (optional) and then run cmake again and make
13:45 Zeno`       if that doesn't work then remove cmake cached files
13:45 Sokomine    did that already.    cmake . -DRUN_IN_PLACE=1 -DENABLE_GETTEXT=1 -DENABLE_FREETYPE=1 -DENABLE_LEVELDB=1 ; make -j2    is my usual compile script. and i even did a make clean once
13:46 Sokomine    hm. gcc 5 does seem to be the first one to have stable support for C++11? and i have 4.9.2? so maybe debian's gcc is too old?
13:46 Zeno`       in source builds... eww
13:47 Fixer       latest mt compiles fine for me
13:47 Fixer       msys2/mingw64/gcc 6.3
13:52 nerzhul     Sokomine, gcc 5.1 is required
13:52 nerzhul     GCC 4.9.2 has partial C++11 support, it doesn't support move constructor (your jsoncpp linking error)
13:52 Sokomine    ah. so no new mt for debian stable users for a couple of days
13:52 nerzhul     you can install GCC 5.1 or 6 or 7 from ubuntu toolchain
13:53 nerzhul     Sokomine, => deb http://mirrors.kernel.org/ubuntu trusty main universe
13:53 nerzhul     and install gcc 6
13:53 nerzhul     and g++ 6
13:54 Sokomine    nerzhul: thanks! will install
13:59 Fixer       Sokomine: or wait before new debian release, it is soon
14:00 nerzhul     41 days
14:00 nerzhul     10
14:00 Sokomine    but...10 days without latest mt? :)
14:01 nerzhul     heh, it's MT with C++11, actually no breakage except compiler :)
14:02 Calinou     Sokomine: try compiling using Clang, maybe
14:02 Calinou     https://packages.debian.org/search?keywords=clang
14:03 Calinou     Jessie has 3.5, which hopefully has full C++11 support
14:03 nerzhul     yeah :D
14:03 nerzhul     in backports No ?
14:03 Calinou     no, in the base repository
14:03 Calinou     Stretch will have 3.8
14:03 Fixer       Sokomine: or use this? https://gitlab.com/nerzhul/minetest/builds/
14:03 Calinou     http://clang.llvm.org/cxx_status.html
14:03 nerzhul     Fixer, heh could be a solution
14:04 Calinou     actually, Jessie even has full C++14, it seems
14:04 nerzhul     Fixer, it's not the official repository
14:04 nerzhul     here is official repository https://gitlab.com/minetest/minetest/pipelines
14:05 nerzhul     debian 9 pipeline failed to download files from time to time locking all pipeline :(
14:07 Fixer       oops
14:09 nerzhul     Calinou, C++14 is marginal in our usage hopefully, C++11 is a huge move C++14 is a must have but not required here heh
14:10 Calinou     yeah, new projects could use C++14 just fine today, MSVC 2017 also has full support for it
14:13 srifqi      Uh, I can't compile Minetest from just fetched repo. ._.
14:14 srifqi      It says an error around mutex. I have no idea what it is.
14:14 Calinou     srifqi: you need a compiler with C++11 support
14:14 Calinou     if you don't have one, just use Minetest 0.4.16
14:14 Calinou     Clang from Ubuntu 14.04 and Debian 8 supports C++11
14:14 Calinou     (but not GCC)
14:15 srifqi      Oh, GCC doesn't support it? That's why.
14:15 Calinou     so you could install Clang, and compile Minetest using it: https://stackoverflow.com/questions/7031126/switching-between-gcc-and-clang-llvm-using-cmake#7032021
14:15 Calinou     recent versions of GCC do, but not the old ones
14:15 Calinou     (Ubuntu 16.04 and Debian 9 are fine)
14:15 srifqi      I use GCC 7.x, iirc.
14:15 Calinou     then it should not be an issue...
14:15 Calinou     try doing a "cmake ." again, "make clean" then "make" again
14:15 srifqi      Maybe I should try to reinstall GCC.
14:16 Calinou     then pastebin the error
14:16 srifqi      The error is too long that Command Prompt window is not enough.
14:17 Calinou     you on Linux or Windows?
14:20 srifqi      On Windows 10, but have Ubuntu on Windows feature. .-.
14:21 Calinou     try "make &> build_log.txt", this will write all command output (including errors) to build_log.txt in the folder where you're compiling Minetest
14:21 Calinou     (works at least on Bash)
14:21 srifqi      Let me try.
14:30 srifqi      Okay, now I get some libraries is missing for cc1.exe to work.
14:30 srifqi      Let me reinstall the compiler.
14:32 srifqi      I was using GCC 7.x from MinGW-W64. Usually I use MinGW from MinGW Installation Manager. Which do you prefer, Calinou?
14:32 Calinou     MinGW-w64 is fine
14:32 Calinou     it doesn't have a package manager like MSYS2 though
14:32 Calinou     (which lets you install commonly-used dependencies easily, thus saving time)
14:33 srifqi      I have MSYS2 installed too.
14:33 srifqi      Shortly, I had 3 compiler: MinGW Installation Manager, MinGW-W64, and from MSYS2.
14:34 srifqi      and now I'm confused.
14:35 Calinou     unfortunately, it's not exactly easy to compile Minetest for Windows... it would be if we provided some sort of "SDK" with all libraries pre-built
14:35 srifqi      But MSYS2 doesn't work with Windows path.
14:35 Calinou     or include the source of dependencies in-tree (but you can already hear the more conservative Linux users yelling at that)
14:35 srifqi      Let me try with my first-installed compiler (from MinGW).
14:36 srifqi      Also, &> is not correct syntax in cmd,
14:37 Calinou     the Windows command promot knows about > and >> IIRC, but not &>
14:37 Calinou     prompt*
14:37 Calinou     not sure if > redirects both stdout and stderr on Windows
14:37 Calinou     srifqi: another thing you could try is compiling for Windows from the WSL ;)
14:37 Calinou     but I've never tried that
14:38 srifqi      What is WSL>
14:38 srifqi      ?*
14:38 Calinou     Windows Subsystem for Linux
14:38 Calinou     "Bash on Ubuntu on Windows" as they call it
14:39 srifqi      Oh, I have that.
14:39 srifqi      So, can I cross-compile?
14:39 red-NaN     or lxss
14:39 red-NaN     srifqi, you should be able to
14:43 srifqi      Calinou, the correct command is `make > log.txt 2>&1`.
14:43 Calinou     yes, it's the POSIX command
14:43 srifqi      red-NaN, have you try it?
14:43 sfan5       make &>log.txt is shorter
14:43 Calinou     but according to https://stackoverflow.com/questions/7526971/how-to-redirect-both-stdout-and-stderr-to-a-file, both work
14:43 Calinou     at least on Linux
14:43 Calinou     (and even on macOS according to some people)
14:44 Calinou     and since the WSL is basically Linux...
14:44 sfan5       Calinou: this "sdk" you want exists, the buildbot
14:44 Calinou     does it include .dll/.lib of all libraries?
14:44 sfan5       every single library is available prebuilt
14:44 Calinou     ah
14:45 sfan5       https://github.com/minetest/minetest/blob/master/util/buildbot/buildwin64.sh#L32
14:45 red-NaN     no but the userland of WSL is identical to ubuntu
14:45 Calinou     yeah, that's what I meant
14:45 red-NaN     as it the binarys are initical
14:46 red-NaN     WSL is implemenet at the kernal level
14:49 srifqi      I was using cmd when I said it.
14:49 srifqi      pastebin or mozilla pastebin?
14:50 red-NaN     doesn't matter
14:50 red-NaN     paste.ee is pretty neat througg
14:51 red-NaN     through*
14:52 srifqi      https://pastebin.mozilla.org/9023740
14:52 srifqi      Calinou
14:53 Calinou     looks like a compiler that doesn't support C++11 to me
14:56 Zeno`       I hope everyone now looks back to my original comments (1-2 years ago) not supporting C++11 and the abusive answers I got directed towards me
14:56 Calinou     Zeno`: well, it's not hard to use a C++11-compliant compiler
14:57 Calinou     but judging of Minetest and MinGW's nature, you're better off cross-compiling from Linux to Windows if you want GCC, or use MSVC on Windows directly
14:57 red-NaN     it's a six year old standard
14:57 srifqi      https://stackoverflow.com/a/25952133/5338238
14:57 Zeno`       Calinou, it kinda is. But anyway, that was ages ago. I am glad to see that my comments/objections were correct though even if it's now causing problems
14:58 red-NaN     srifqi, what compiler are you using?
14:58 Calinou     yes, you need to install with POSIX threads, not win32 threads
14:59 srifqi      Calinou, that's why. .-.
14:59 srifqi      red-NaN GCC 5.3.0
14:59 Calinou     pretty sure the default is to use POSIX threads :)
14:59 Calinou     (in the installer)
15:00 red-NaN     huh you can crosscompile minetestw ith gcc?
15:01 Zeno`       At the time when C++11 was first raised (when there were still many many LTS linux distros that did not support it) I was accused of objecting to C++11 because I was too old ROFLOL
15:01 red-NaN     now it's just debain (tm)
15:02 Zeno`       red-NaN, yeah. But I am vindicated finally
15:02 Calinou     you could use Clang on Debian 8 and Ubuntu 14.04, and it'd work as usual
15:03 Calinou     Clang is not an obscure compiler anymore, it works well for lots of things now :P
15:03 red-NaN     Calinou, would appimages solve this?
15:03 Calinou     red-NaN: AppImages would allow you to *run* Minetest on these systems, without compiling
15:03 Calinou     but not to build yourself...
15:03 nerzhul     Zeno`, i want to abuse you by asking you to help me to merge some PR :p
15:04 Calinou     (unless you're willing to use Clang, or find a custom repository with up-to-date GCC)
15:04 red-NaN     well if we added daily appimage builds...
15:04 Zeno`       nerzhul, but I am watching Star Trek
15:04 nerzhul     i's not incompatible with C++11 :p
15:04 red-NaN     that would solve a lot of the reasons for which people want to build minetest
15:05 glut32      hi all
15:05 Calinou     maybe they have GCC 8 and Clang 5 in Star Trek ;)
15:06 nerzhul     oh C++11 is obsoleted in star trek and doesnt' compile, understood
15:08 glut32      My job mate has this error : 2017-06-07 16:30:37: WARNING[Main]: MapBlock::deSerialize(): Ignoring an error while deserializing node metadata at ((0,2,19): deSerializeLongString: size not read
15:08 glut32      and after a bad_alloc
15:08 glut32      2017-06-07 16:30:37: ERROR[MeshUpdate]: An unhandled exception occurred: std::bad_alloc
15:10 Calinou     glut32: what Minetest version is the client/server on? and operating system?
15:11 glut32      this is where the execption : https://github.com/minetest/minetest/blob/master/src/util/serialize.cpp#L218
15:11 glut32      He used win7
15:11 glut32      on a custom map
15:12 red-NaN     does it happen every time the map is loaded?
15:12 glut32      Do you think that the map is corrupted
15:12 glut32      only after 5 or 10 minutes
15:13 red-NaN     looks like a damaged mapblock
15:13 red-NaN     but it shouldn't crash minetest
15:13 glut32      I think that size is not setted and cuz a bad_alloc (no memory available)
15:14 glut32      *cause
15:14 nerzhul     bad_alloc no memory means your PC has no more memory for the problam
15:14 nerzhul     program
15:14 sfan5       because it throws an exception it should never come to allocate any memory
15:15 glut32      but there memory available like 1go
15:15 glut32      we see the memory usage
15:16 glut32      the game consumes 1.7 go
15:16 glut32      and we are on a 6go ram pc
15:19 Fixer       use 64 bit minetest
15:21 Fixer       32 bit one has tendency to run out of memory if you explore big areas
15:25 Zeno`       There is no memory allocation on that line :/
15:27 kilbith     Zeno`, just curious, what are those "abusive answers"?
15:28 Zeno`       kilbith, you know exactly what they were
15:28 kilbith     no? I don't sorry
15:28 Zeno`       The comments made by gcu when I objected at the time to moving to C++11
15:29 kilbith     where they are?
15:29 kilbith     oh on the forum
15:29 Zeno`       IRC logs. Probably on the forum as well.
15:29 nerzhul     2 years ago why not, all compilers were not ready heh
15:29 nerzhul     now all projects are moving because they were waiting for redhat and debian
15:30 kilbith     Zeno`, dug it out: https://forum.minetest.net/viewtopic.php?p=169046#p169046
15:30 kilbith     indeed, abusive...
15:30 Zeno`       lol, that is not what was said on IRC :P
15:31 nerzhul     jsoncpp, yamlcpp, minetest, SDL2PP, libreoffice etc :D
15:31 kilbith     Zeno`, what was said on IRC?
15:32 Zeno`       Search the logs previous to that date. You can use gregorycu, zeno and sfan5 as keywords
15:34 kilbith     I'll probably ask gcu himself
15:34 Zeno`       I'm sure you'll be honest enough to post the link (that you know exists) once you find it
15:34 kilbith     it's fair to get both parties versions of the history
15:34 Zeno`       it's all in the logs
15:34 kilbith     of course
15:36 * srifqi    is trying to remake Minetest after reconfiguring it.
15:38 srifqi      Look! Different error.
15:40 kilbith     Zeno`, http://irc.minetest.ru/minetest-dev/2015-01-29#i_4135010
15:40 sfan5       ???
15:40 sfan5       how does personal drama keep leaking here
15:40 sfan5       this is the dev channel
15:40 kilbith     adressing Zeno' request
15:41 Zeno`       That is not the conversation I am referring to
15:41 kilbith     let's move it to -hub, ok?
15:42 srifqi      C:/mingw-w64/mingw32/i686-w64-mingw32/include/crtdefs.h:123:1: error: '__MINGW_EXTENSION' does not name a type; did you mean '__GNUC_EXTENSION'?
15:43 srifqi      .-.
15:57 red-NaN     #5852
15:57 ShadowBot   https://github.com/minetest/minetest/issues/5852 -- Improve the path select GUI by red-001
15:57 red-NaN     could anyone review it?
15:58 red-NaN     it's a useful improvement and bugfix
17:05 nerzhul     merging #5942 in ~10 mins
17:05 ShadowBot   https://github.com/minetest/minetest/issues/5942 -- make ret variable in /builtin/mainmenu/tab_credits.lua local by DS-Minetest
18:20 itubal      Hi all
19:30 red-NaN     could someone restart https://travis-ci.org/minetest/minetest/jobs/240491344 ?
19:34 sfan5       done
19:36 red-NaN     that build seems to randomly fail fairly often
19:41 nerzhul     merging #5940 in ~10 mins
19:41 ShadowBot   https://github.com/minetest/minetest/issues/5940 -- Use a settings object when generating world.mt and set player_backend to sqlite. by red-001
19:44 nerzhul     sfan5, can you take 10 mins to test/approve #5928 please ? :)
19:44 ShadowBot   https://github.com/minetest/minetest/issues/5928 -- C++11 patchset 5: use std::threads and remove old compat layer by nerzhul
19:45 red-NaN     !tell Krock are you ok with #5852 now?
19:45 ShadowBot   red-NaN: O.K.
19:46 nerzhul     red-NaN, please fix file into the clang-format-whistelist please to make lint okay
19:46 nerzhul     you renamed it and it's not in the whitelist either :)
19:47 nerzhul     i think we should configure clang-format to use 1 line per variable instead of this package mode no ? sfan5 ? => https://travis-ci.org/minetest/minetest/jobs/240492414
19:47 sfan5       what do you mean?
19:48 nerzhul     i mean in the following link i can configure clang-format to cofmrat using the --- lines instead of +++ lines :)
19:48 nerzhul     to format*
19:48 nerzhul     1 variable per line in constructor instead of packaged mode based on line length
19:49 nerzhul     many discuss about this on meeting and then define a code style rule and apply it with clang-format ?
19:49 sfan5       if that's possible it would be nice to have it
19:51 nerzhul     sounds to be ConstructorInitializerAllOnOneLineOrOnePerLine, testing it
19:51 nerzhul     https://clang.llvm.org/docs/ClangFormatStyleOptions.html
19:54 nerzhul     yeah it works well
19:54 nerzhul     if initializer fit on one line, 1 line else 1 per line
19:54 nerzhul     sounds nice
20:07 nerzhul     mm the rule is nice except the colon... and i don't see how to control it
20:12 nerzhul     here is the result, i'm not very fan: https://github.com/minetest/minetest/pull/5944
20:13 rubenwardy  heh
20:13 rubenwardy  I prefer          ClassName(params):
20:13 rubenwardy  which is what most of the code consistently uses
20:15 kilbith     that guy should be more listened: https://github.com/minetest/minetest_game/issues/1757#issuecomment-306745450
20:15 kilbith     a real truthful asshole who often cut through all the BS
20:15 kilbith     I love that
20:16 nerzhul     rubenwardy, i prefer it too but there is no option, then... fuck :p
20:16 nerzhul     closing the PR i don't see how to solve it properly except sending upstream patch :p
20:17 red-NaN     kilbith, like you said code talks bullshit walks
20:17 nerzhul     red-NaN, https://github.com/minetest/minetest/pull/5852/commits/c41e98a889ba187464d55ea04a87deac6f75000d
20:17 nerzhul     seems you paste a link in the whitelist :p
20:17 red-NaN     I have no idea how that even happened
20:18 red-NaN     copy and paste seems to be working in myterious ways lately
20:18 nerzhul     libevent2 is very nice to handle low level socket using event API and portable on all OS i think it's a good choice to rewrite the network stack properly, it's robust and widely used
20:22 nerzhul     (i'm starting to poc a new stack which is efficient and lightweight)
21:12 itubal      Hi
21:13 itubal      good night everybody
21:13 itubal      I want to make a question
21:14 itubal      I'm doing a craftnode. whow can I put some of this nodes in my world??
21:15 itubal      Sorry
21:15 itubal      i put craftnode.... CRAFTITEM is the correct term
21:16 calcul0n    itubal, make a recipe so players can craft it ?
21:17 calcul0n    or do you mean a node which should be present on the map ?
21:31 itubal      I try to put the craftitem with :
21:31 itubal      minetest.add_item
21:32 itubal      first i need de craftitem
21:34 itubal      https://rubenwardy.com/minetest_modding_book/chapters/nodes_items_crafting.html
21:35 itubal      The same a "diamonds_fragments"
21:36 itubal      No, it isn't a node, a basic node.
21:37 itubal      it is a "craftitem"
21:38 calcul0n    there is no add_item method
21:39 calcul0n    Craftitems cannot be placed in the world.
21:39 calcul0n    i don't see what you are trying to do
21:39 itubal      I can see the craftitem in my inventory
21:40 itubal      i want to see this "node" this crafftitem in my world
21:40 calcul0n    so you have to register it with register_node instead of craftitem
21:41 calcul0n    add use minetest.set_node()
21:41 itubal      no. I dont want a node. I want a craft item
21:43 itubal      https://github.com/itubal/ejemplosminetest
21:43 itubal      this is the source code
21:44 calcul0n    damn, i was wrong, there is an add_item method :)
21:44 itubal      i want a craftitem. The player gets this craftitems and with them make a craft
21:45 itubal      yes there is a add_item
21:46 itubal      please download the repository
21:46 itubal      https://github.com/itubal/ejemplosminetest
21:46 itubal      do you speak spanish?
21:46 calcul0n    no, sorry
21:47 calcul0n    but it looks like add_item takes either an ItemStack or a table like { name="itemname" }
21:47 itubal      o don't worry. i'n making some mistakes ... but i'm sure that you can understand me
21:47 calcul0n    yes i do
21:48 calcul0n    maybe try something like minetest.add_item({x=4, y=1, z=3}, { name="mimod:esparto" })
21:50 calcul0n    ho, a simple string seems to be fine too after all
21:51 itubal      I'll try
21:51 itubal      {name = "mimod:esparto"}
21:53 itubal      I copied this line from a scripts that is working
21:53 calcul0n    yes, i also see some mods using a simple string, so it shouldn't be the problem
21:54 calcul0n    are you sure the y positions are above the ground ? how did you get them ?
21:58 itubal      In the game.
21:59 itubal      i can see the position pressing F5
22:00 itubal      i'm trying this ... and ... doesn't appear
22:01 calcul0n    maybe you should add 1 to all y positions
22:01 calcul0n    not sure, but they are maybe spawning in the ground
22:02 itubal      2017-06-07 23:59:13: WARNING[Main]: Irrlicht: PNG warning: iCCP: known incorrect sRGB profile
22:02 calcul0n    i always have this warning, i don't think it's related
22:03 itubal      ok ... this is usefull to know