Minetest logo

IRC log for #minetest-dev, 2017-06-04

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:05 Sokomine sfan5: the full backtrace is 1.5 mb in size and seems to be too long for github. any hints how to best provide it?
00:08 Taoki joined #minetest-dev
00:10 ShadowNinja Sokomine: The regular backtrace and the full backtrace with just the first two or three functions should be enough.
00:18 ShadowNinja Looks like propagateSunlight assumes that the layer above pmax is in the vm.
00:19 ShadowNinja paramat is probably most qualified to fix this.
01:36 IhrFussel joined #minetest-dev
01:36 IhrFussel left #minetest-dev
02:13 GreenDimond joined #minetest-dev
02:14 GreenDimond So... why weren't the new sandstones included in 0.4.16???
02:14 GreenDimond Oh wait
02:14 GreenDimond o_O
02:15 GreenDimond >~< Minetest didnt upgrade properly
04:49 AntumDeluge joined #minetest-dev
05:07 Raven262 joined #minetest-dev
05:34 nore joined #minetest-dev
05:47 Thomas-S joined #minetest-dev
06:14 cx384 joined #minetest-dev
06:58 Hunterz joined #minetest-dev
07:16 cx384 joined #minetest-dev
07:28 nerzhul joined #minetest-dev
07:31 nerzhul hi, merging #5827 in ~1à mins
07:31 ShadowBot https://github.com/minetest/minetest/issues/5827 -- Mapgen v6: Use snow blocks in tundra and remove them from taiga by CasimirKaPazi
07:37 nerzhul i removed version changes from #5814 as it's not required in this PR, will merge it in ~20 mins as now it's okay why everybody
07:37 ShadowBot https://github.com/minetest/minetest/issues/5814 -- Minetest for C++11 (CMakeLists + Travis) by nerzhul
08:26 Krock joined #minetest-dev
08:29 Krock cl : Command line warning D9002 : ignoring unknown option '/std:c++11'   but I'm surprised that the new JsonCPP ships compatibility code
08:30 sfan5 tell nerzhul
08:30 Krock nah, it's because C++11 isn't supported here. I wonder how long it will take until I can't compile Minetest anymore :P
08:31 sfan5 not long because non c++11 is still supported
08:32 sfan5 though as soon as #5821 is merged non-c++11 will no longer work
08:32 ShadowBot https://github.com/minetest/minetest/issues/5821 -- C++11 patchset 2: remove util/cpp11.h and util/cpp11_container.h by nerzhul
08:34 Krock hehe, the irony is that it still will compile after this one
08:35 Krock since MSVC 2010 already supports unordered sets/maps
08:35 Krock ah, he requested a review from me.. testing
08:39 Krock lol, it fails to compile because constexpr is no longer defined. *adds glue*
08:40 sfan5 it might be time to update your compiler
08:40 Krock nah, time to update my OS
09:00 DS-minetest joined #minetest-dev
09:00 lisac joined #minetest-dev
09:11 YuGiOhJCJ joined #minetest-dev
10:08 Fixer joined #minetest-dev
11:41 kilbith joined #minetest-dev
11:51 fwhcat joined #minetest-dev
12:02 Shara joined #minetest-dev
12:19 red-001 could someone review #5852?
12:19 ShadowBot https://github.com/minetest/minetest/issues/5852 -- Improve the path select GUI by red-001
12:23 Hunterz joined #minetest-dev
12:46 CalebDavis joined #minetest-dev
12:57 kilbith_ joined #minetest-dev
13:53 red-001 some users of the android app seems to be reporting not being able to place blocks
13:59 kilbith joined #minetest-dev
14:02 behalebabo joined #minetest-dev
14:14 DS-minetest joined #minetest-dev
14:17 iZacZip joined #minetest-dev
14:20 EDAKIRI joined #minetest-dev
14:23 paramat joined #minetest-dev
14:46 octacian joined #minetest-dev
14:46 SG9kb3I joined #minetest-dev
14:49 paramat game#1749 game#1758
14:49 ShadowBot https://github.com/minetest/minetest_game/issues/1749 -- Flowers: New texture for yellow dandelion by paramat
14:49 ShadowBot https://github.com/minetest/minetest_game/issues/1758 -- Stairs: Use one recipe matching inventory appearence by paramat
14:49 paramat good
14:50 rubenwardy huh
14:50 rubenwardy https://forum.minetest.net/viewtopic.php?f=3&amp;t=17789   shouldn't be in general discussion, and shouldn't be a global announcement :P
14:56 kilbith SG9kb3I: because it's you
14:59 Fixer heh, someone noted on forum that mine_test_ - test part is giving bad impression :p
15:00 Fixer can't agree more
15:00 fwhcat but minestone would make wood fanatics cry
15:03 Fixer just replace that test part
15:03 Fixer it is not alpha or smth
15:03 Fixer more like beta :}
15:03 Fixer very stable beta
15:03 Fixer pre 1.0
15:03 kilbith pointless discussion
15:03 kilbith code talks, bullshit walks
15:04 red-001 well thats some advance bullshit
15:04 Fixer advanced*
15:04 red-001 most just sits there and doesn't do anything
15:07 kilbith SG9kb3I, I only see whitespaces
15:08 red-001 kilbith, it's hodor not hunter2
15:08 kilbith well now it's clear it's Zeno
15:09 Thomas-S I think that #5895 is ready for review.
15:09 ShadowBot https://github.com/minetest/minetest/issues/5895 -- Meta-set nodedef: Part 1 by Thomas--S
15:24 Calinou latest Minetest Git does not seem to build due to libjsoncpp errors, on Ubuntu 14.04
15:24 Calinou (I am building AppImages right now)
15:26 Krock using C++11?
15:27 Calinou I didn't touch the cmake settings
15:27 Calinou I did re-run "cmake ." though
15:28 Calinou https://hastebin.com/luyofuxexi.txt
15:28 Calinou "Using bundled JSONCPP library."
15:28 Calinou can I force the system one to fix this? I have it installed
15:29 Calinou found the option for it, let's try
15:29 Calinou (it's ENABLE_SYSTEM_JSONCPP=1)
15:31 Calinou it compiled just fine, what about using system jsoncpp if it's found by default? unless other distros cause issues with it
15:31 Calinou server seems to run fine
15:31 Calinou (but maybe it doesn't depend on it)
15:31 Calinou can't test client right now, this is an headless VM
15:34 Grandolf joined #minetest-dev
15:34 halt_ joined #minetest-dev
16:29 fwhcat_ joined #minetest-dev
16:37 nerzhul joined #minetest-dev
16:57 kilbith joined #minetest-dev
17:09 Player_2 joined #minetest-dev
17:26 YuGiOhJCJ joined #minetest-dev
17:35 kilbith joined #minetest-dev
17:46 MoNTE48 joined #minetest-dev
18:04 nerzhul merging #5821 in ~1h
18:04 ShadowBot https://github.com/minetest/minetest/issues/5821 -- C++11 patchset 2: remove util/cpp11.h and util/cpp11_container.h by nerzhul
18:12 garywhite wait, you guys are already adding fixes to 0.4.16?
18:12 garywhite Do you guys ever take a day off?
18:18 Krock how about no?
18:18 VanessaE ^^^ that is the start of the 0.5.x branch
18:19 VanessaE there are lots of PRs that didn't make the deadline for 0.4.16
18:19 VanessaE I just hope stuff will be backported to 0.4.x for point releases later
18:19 Krock nerzhul came up with the idea of a parallel 0.4.17 release to backport all bugfixes for the 0.4.x line
18:19 VanessaE ninja'd
18:19 Krock yeah
18:20 VanessaE slowpoke ;)
18:23 VanessaE good G*d, speaking of slowpokes, there are still three 0.4.12 servers on the list
18:23 CalebDavis joined #minetest-dev
18:24 Krock as long there are no exploitable worldedit_gui versions around anymore it's not too horrifying
18:24 VanessaE heh
18:25 halt_ joined #minetest-dev
18:31 octacian Any thoughts as to how modmgr could determine whether a mod is a normal server-side mod or a clientmod?
18:32 Krock using minetest.get_modpath? dunno how else
18:33 octacian Krock: Nope, this is for installing mods. I need to know whether to put the mod in /mods or /clientmods
18:33 octacian Best thing I can come up with is either just specify it as a parameter to the function (meaning that clientmods and normal mods have to be installed separately in the formspec) or to ask clientmod creators to add something to their mod.conf
18:33 octacian (neither of which being very nice IMO)
18:33 Krock or another setting in mod.conf ?
18:33 octacian Any thoughts?
18:33 octacian Yeah, maybe `type = "client"`?
18:34 octacian Maybe get it posted in a pinned topic in clientmods subforum then, that all clientmods should have that set
18:34 Krock without "" but yes, that's about what I imagined
18:34 octacian Right... heh, Always forget that
18:34 octacian I'll go for that method for now, seems best
18:35 octacian (I'm rewriting modmgr - it REALLY needs it)
18:47 paramat i rarely take days off MT, i worked christmas day too :]
18:49 Krock this->family_events.clear();
18:49 Krock delete this->family_events;
18:49 Krock this->family_events = NULL,
18:49 Krock err, I messed it up. Swap '.' with '->'
18:58 nerzhul VanessaE, 0.4.12 servers are always supported, i hope they will migrate to more recent version since next release because we keep network compat for 2 years (except this 0.5.0 which will do a break, and the 0.4.X will just die after some time
18:59 nerzhul yes i have idea we can make a last 0.4.17 with 0.5.0, 0.4.17 containing only backportable fixes
18:59 nerzhul we have the stable 0.4 for this
18:59 VanessaE nerzhul: I was making a jab at pokey admins who are basically pulling another RedCrab ;)
19:00 ShadowNinja nerzhul: looks like your JSONCPP update broke the Launchpad builds.
19:00 nerzhul !tell ShadowNinja can you add a point for next meeting to discuss about a 0.4.17 bugfix version released at a moment
19:00 ShadowBot nerzhul: O.K.
19:00 nerzhul ShadowNinja, use C++11 compiler
19:01 nerzhul jsoncpp 1.8 is for C++11 only
19:02 ShadowNinja nerzhul: I'm not sure what needs to change.  Build log: https://launchpadlibrarian.net/322492256/buildlog_ubuntu-trusty-amd64.minetest_201706041832-0~2362d3f~ubuntu14.04.1_BUILDING.txt.gz
19:03 nerzhul ShadowNinja, let me a day or two to fix it on gitlab.com 14.04 docker build
19:03 nerzhul i need to fix gitlab pipeline for debian/ubuntu daily build i started to produce, i will see what we should improve
19:03 Hijiri since code freeze is over can someone review #5612?
19:03 ShadowBot https://github.com/minetest/minetest/issues/5612 -- Allow overriding tool capabilities through itemstack metadata by raymoo
19:04 Hijiri or do I need to rewrite things to use C++11 features
19:04 Hijiri or maybe I will have to rebase when everything is changed to use C++!1
19:04 nerzhul Hijiri, no need i think
19:04 nerzhul just rebase it should not affect your pr
19:05 Hijiri ok
19:05 nerzhul except if you use old compat macro we removed today
19:05 nerzhul (UNORDERED_MAP, UNORDERED_SET)
19:06 nerzhul looking at your PR i don't see any C++11 affected line or C++11 improvement to add to your PR
19:06 nerzhul but rebasing is good, you use jsoncpp and jsoncpp was updated
19:08 behalebabo joined #minetest-dev
19:10 Krock merging #5891 in 20 mins
19:10 ShadowBot https://github.com/minetest/minetest/issues/5891 -- Remove deprecated code segments by Thomas--S
19:14 paramat joined #minetest-dev
19:19 Hijiri nerzhul: ok, I rebased, also tested and the feature still seems to work
19:19 Hijiri wait no
19:19 Hijiri maybe I didn't rebase properly, it's not showind differences
19:20 Hijiri yeah I forgot to actually rebase
19:23 Hijiri I did get some errors, I think because I used UNORDERED_MAP
19:26 kilbith joined #minetest-dev
19:27 behaleba- joined #minetest-dev
19:58 Raven262 joined #minetest-dev
20:03 nerzhul Hijiri, okay, just fix them
20:08 bigfoot547 joined #minetest-dev
20:10 Grandolf joined #minetest-dev
20:10 Hijiri nerzhul: I did already
20:13 CalebDavis since the move code was rewriten will 0.4.16 soon give way to 0.4.17
20:13 CalebDavis due to bugfixes
20:14 nerzhul why we will release a 0.4.17 just before 0.4.16 ?
20:14 CalebDavis no i was asking if 0.4.16 would be a short term release
20:15 CalebDavis cause of the major updates that will need to be updated due to enough bugfixes
20:15 paramat no, 0.5.0 next in winter
20:15 CalebDavis ok cool
20:16 CalebDavis so that is fairly short term
20:16 paramat well, normal interval
20:16 CalebDavis ok i wasnt around when 0.4.14 updated to 0.4.15 XD
20:16 nerzhul 0.5.0 keep same interval, but maybe we will do a 0.4.17 patch release in stable-0.4 branch with bugfix backports, we will discuss about this in a next meeting
20:17 paramat yeah ^
20:17 CalebDavis ok thx that was what i was wondering
20:25 Tmanyo joined #minetest-dev
20:25 Grandolf left #minetest-dev
20:38 nerzhul wow gdb 8 was released and also decided to switch to cpp11 :)
20:40 nyuszika7h_ joined #minetest-dev
20:41 behaleba- joined #minetest-dev
20:45 Ritchie_ joined #minetest-dev
20:45 rubywarden joined #minetest-dev
20:48 Karazhan joined #minetest-dev
20:57 paramat joined #minetest-dev
21:03 fireglow[m] joined #minetest-dev
21:06 proller joined #minetest-dev
21:07 silwol joined #minetest-dev
21:15 nerzhul wow that's very elegant i didn't meant to use this model: https://twitter.com/mrkkrj/status/854241747936915456
21:18 paramat is it our convention to use 'm_' prefix for private member variables only?
21:19 nerzhul we use m_* for private or protected members yes, why ?
21:20 paramat ok, just wondering the usage
21:20 juhdanad joined #minetest-dev
21:22 juhdanad Hi, everybody!
21:22 nerzhul it's a global usage in CPP/Java you have generally two models: m_flat_case (generally C++) or mCamelCase (java)
21:22 paramat i'm now working on a PR to limit player spawn to within mapgen edges, it uses 'calcMapgenEdges()' to get exact values
21:23 paramat with a very small 'mapgen limit' 'findSpawnPos()' tends to spawn a player out in the void :]
21:23 juhdanad Congratulations for the release! (a bit late, I know)
21:23 paramat hi
21:25 nerzhul juhdanad, ty
21:26 EDAKIRI joined #minetest-dev
21:27 juhdanad This is the second release I took part in, and I learned an important lesson: if you create a lot of bugs with your PRs, then you can create a lot of bugfixes and the net result is many merged commits.
21:33 Krock or you're a weblate translator and cheat commits this way
21:33 nerzhul i squash them :p
21:34 Krock all but drunk russians
21:34 nerzhul juhdanad, to be honest the part you modified was a old part and is very sensible :p
21:34 nerzhul Krock, yes, if you cheat with 2 accounts i'm lost :(
21:34 juhdanad I will try to fix that bug with mapgen lighting...
21:35 paramat nice
21:35 paramat i guess for a node above the vm you can only guess according to > or < water_level?
21:36 juhdanad The mapgen can only guess. A voxel manipulator can load the block above.
21:36 nerzhul paramat, now we use C++11 i think we can make many problematic settings thread_local isntead of static and if i remember the caller thread lifetime is the game lifetime :)
21:37 paramat ah
21:37 nerzhul then this will fix all settings not changed if mt is not restarted if i'm exact heh
21:37 paramat yeah load above then guess if necessary
21:37 paramat water_level is used in MT as the dividing level between assuming light and assuming dark
21:39 paramat i will also fix player respawn, using the SAO pos check code
21:41 juhdanad Is there any way to connect to this channel on the http port? On weekdays I have a connection which does not allow IRC.
21:41 paramat because 'objectpos over limit()' is inaccurate
21:41 juhdanad Previously I connected with kiwiirc.
21:41 paramat erm
21:42 nerzhul juhdanad, webchat.freenode.net
21:42 paramat isn't webchat blocked here though?
21:42 nerzhul no i'm using it time to time
21:43 paramat works for dev channel?
21:43 nerzhul y
21:43 juhdanad nerzhul: thank you! Then I will be able to join there on weekdays too!
21:49 red-001 we definatly banned webchat
21:49 red-001 *!*@gateway/web/freenode/* on Sat May 13 17:43:48 2017 by ShadowNinja!~owen@unaffiliated/shadowninja
21:52 nerzhul oh
21:52 nerzhul but
21:52 nerzhul i did it yesterday
21:53 nerzhul :p
21:53 nerzhul (maybe because i'm operator & auth)
21:54 juhdanad joined #minetest-dev
21:54 red-001 maybe it would be possible to set up an exemption for juhdanad then?
21:54 nerzhul works nice :p
21:54 red-001 juhdanad, could you try and say something?
21:55 juhdanad joined #minetest-dev
21:55 juhdanad Could not send text, only read.
22:05 nerzhul i'm pushing a gitlab-ci.yml fix permitting to Debian 8/Ubuntu 14.04 build correctly on daily builds
22:06 Miner_48er joined #minetest-dev
22:06 Fixer nerzhul: ty
22:07 Fixer nerzhul: meanwhile i'm trying to compile it via msys2/mingw64 under windows, "i have no idea what i'm doing" (c)
22:08 nerzhul Fixer, it doesn't fix the ubuntu PPA, i don't know how to fix it, but if someone want to fix it, just use toolchain and GCC6 from trusty
22:08 Fixer nerzhul: i will wait two weeks for debian 9 :}
22:08 nerzhul See here: https://github.com/minetest/minetest/commit/bfacfc20624a46f05193e7e036a58ec906450494#diff-96edf7a6f008de9e928d04e1ae5e12a5R123
22:08 nerzhul for debian 8 you can do the same
22:08 nerzhul look at the ^ diff
22:08 nerzhul https://github.com/minetest/minetest/commit/bfacfc20624a46f05193e7e036a58ec906450494#diff-96edf7a6f008de9e928d04e1ae5e12a5R67
22:09 Fixer o_o
22:09 nerzhul pipeline is in three steps, build, package, deploy
22:09 nerzhul package and deploy doesn't have any build tools installed
22:10 juhdanad I have successfully reproduced the crash!
22:10 Fixer cool
22:10 Fixer nerzhul: you are using ubuntu ppa on debian?
22:10 Devy joined #minetest-dev
22:10 nerzhul trusty and debian 8 are very close versions
22:10 nerzhul trusty is based on debian 8 if i remember
22:11 nerzhul then you can use it safely, as you see it works on the debian 8 docker image properly
22:12 nerzhul you can test it :)
22:12 nerzhul gcc packages are just installing gcc and doesn't have hard deps, it should work properly without breaking system
22:12 nerzhul i should go, see you
22:14 juhdanad The strange thing is that my OS sliently ignored the invalid array deference, and the game worked. I had to insert an assert to find the bug.
22:15 Devy Hello, when compiling the source code from this page: https://github.com/minetest/minetest/releases for the new release, and at the same time following the instructions for cloning from git, an error occurs that says error: ‘::max_align_t’ has not been declared. This is in the file: /minetest/scr/util/srp.cpp. Adding this line "#include <cstddef>" allows it to complie. This is on Linux Ubuntu 14.04
22:17 red-001 Devy, do you have a github account?
22:17 Devy Yes
22:17 red-001 then feel free to make a PR
22:18 red-001 chances are it will get merged pretty fast
22:19 Devy Ok, but I'm not all too familiar with c++ and there is some code in there that has to do with cross-platform stuff, so i want to make sure who ever adds it knows what they are doing, and that is not me.
22:20 red-001 well at least create an issue on github
22:21 red-001 something thats just mentioned on IRC will likely be forgotten
22:21 juhdanad paramat: mapgen v6 calculates lighting in a larger volume. Why is that?
22:22 paramat erm
22:22 Fixer Msys2/mingw64 build was succesful
22:23 paramat maybe because mudflow, grass and tree addition occurs within the shell too
22:23 paramat out to fullmin/max
22:25 Devy There, I put up a pull request.
22:26 Devy https://github.com/minetest/minetest/pull/5910
22:26 paramat #5908
22:26 ShadowBot https://github.com/minetest/minetest/issues/5908 -- (Re)spawn players within mapgen edges by paramat
22:26 red-001 #5910
22:26 ShadowBot https://github.com/minetest/minetest/issues/5910 -- Error when compiling from source fix. by CoderForTheBetter
22:27 Fixer i have strange news
22:28 Fixer msys2/mingw64 build has sooo much less stutter
22:29 Fixer i'm quite amazed by difference
22:30 red-001 oh great our windows using dev disappres just when you need them
22:30 red-001 windows-using*
22:31 Fixer to run msys2/mingw64 i've used three files shipped with mingw: libstdc++-6.dll, libgcc_s_seh-1.dll, libwinpthread-1.dll
22:32 Fixer too good to be true
22:33 red-001 oh
22:33 * red-001 feels like an idiot
22:33 red-001 I forgot to copy the needed dll's
22:33 red-001 that most be why it was crashing for me
22:35 Devy left #minetest-dev
22:37 paramat yes if it's possible to set up a webchat exemption for juhdanad we will, not sure it will be
22:39 juhdanad paramat: thank you! (I addressed #5900)
22:39 ShadowBot https://github.com/minetest/minetest/issues/5900 -- segfault in mapnode.h; crash
22:40 red-001 apprently <kunwon1> /mode #channel +e nick!user@host
22:40 red-001 give juhdanad voice should also work
22:40 red-001 giving*
22:44 srifqi joined #minetest-dev
22:54 Fritigern joined #minetest-dev
23:03 QwertyDragon_ joined #minetest-dev
23:05 octacian #5912
23:05 ShadowBot https://github.com/minetest/minetest/issues/5912 -- Main Menu: Add get_clientmodpath API by octacian
23:19 Devy joined #minetest-dev
23:19 paramat rubenwardy i wonder if you could OP me so i can voice juhdanad sometime?
23:22 paramat thanks
23:22 paramat useful anyway as i'm around during the EU night
23:30 Fixer joined #minetest-dev
23:36 Taoki joined #minetest-dev
23:47 Devy paramat, can I talk to you about lua code here?
23:48 paramat about a mod?
23:48 Devy Yes. I talked to you a few days ago through the forums.
23:49 paramat on minetest channel
23:49 Devy Ok.

| Channels | #minetest-dev index | Today | | Google Search | Plaintext