Time Nick Message 00:18 PilzAdam is anyone against this? https://github.com/PilzAdam/minetest/commits/modsystem 00:18 Jordach nope.avi 00:19 Exio i actually like the current "way" 00:32 PilzAdam Exio, what dont you like in the new system? 00:32 Exio having to edit a file instead of just "cp" 00:33 Exio or how are new mods (not in the list) managed? 00:33 iqualfragile they are ignored 00:34 PilzAdam yea, now you need to: cp && echo "load_mod_$(name) = true" >> world.mt 00:34 iqualfragile but otherwise it would get reeealy hard to write a modmanager 00:34 PilzAdam its just better to disable it by default to not mess up other worlds 00:35 Exio hm 00:37 PilzAdam we already kinda have decided to do it like this, now only the code needs to be reviewed 00:41 PilzAdam Ill guess I merge it tomorrow 00:41 PilzAdam -ll 00:43 hmmmm back 00:43 Exio wb 00:44 hmmmm anyway, doesn't seem like the texture visual preloading is more than 200ms slower with the menu drawing 00:44 hmmmm it's just that i wasn't used to seeing the progress like that and it seemed slower 00:44 hmmmm also it was about a second slower because i was watching a video at the time 00:44 Exio hehe 00:45 hmmmm as for the stuttering, i'm still looking at that 00:47 hmmmm i probably shouldn't bother... the last time i tried to track down a stuttering regression i didn't manage to find it 01:05 sapier1 ok abusing irrlicht tab as buttons seams to work 01:06 Jordach seems* 01:06 sapier1 VanessaE could you check if new menu size fits your requirements? (size lock is still missing so plz window mode only) game and modmanager are not adjusted 01:06 sapier1 grr ... I wonder how often I'm gonna make this mistake again 03:26 hmmmm anyway 03:26 hmmmm https://github.com/kwolekr/minetest/commit/45efa66e54a0db558b97ec8326a5827ed5819816 03:26 hmmmm about time i got this done 03:27 hmmmm before i push this to upstream, i need somebody who gets the warnings to test it and see if there are any more to deal with 03:32 kahrl btw delete NULL is a no-op so the ifs in ~Ore are redundant 03:50 hmmmm i see 03:54 kaeza VanessaE, around? 03:54 kaeza woops wrong channel. sorry 07:10 hmmmm yeah... it might be something with the meshmaking, i don't know 07:11 hmmmm at c824abd7b2e199d9c7e3dcc2839a28275f4120f2, map starts emerging immediately and i don't notice as significant stuttering 07:17 hmmmm and at 2c09e8a84cac653b85bba3c85d781e24b5d6f4f8 the problems are there 07:18 hmmmm also i've noticed that when i enter a game on this commit, i sometimes get a crash 07:18 hmmmm it seems random, and it's in GUIChatConsole::GUIChatConsole 07:28 hmmmm i can't really figure it out, but it seems like m_font->getDimension(L"M"); is segfaulting with an address of 0... perhaps getDimension() is the thing that's NULL? 07:28 hmmmm i think my debugging environment might be screwy, but that's what it says from the watch anyway, and i can't see the disassembly because it won't let me jump to rip for whatever reason... 07:32 hmmmm indeed, it crashed at call qword [rax + 0x08], and it seems like getDimension is the 3rd virtual function in IGUIFont 07:33 hmmmm is this the infamous F10 bug?! 07:35 hmmmm rax was 1 at the time of the crash, not 0.. 07:36 hmmmm i don't really know where this heap corruption might be happening 07:36 hmmmm later 12:34 kahrl hmmmm: I believe the segfault you were talking about was fixed in 587e7b299b 13:24 sapier ok another best version ever https://github.com/sapier/minetest/tree/next_gen_main_menu ... i18n is still missing ... imho we need a generic solution available to mods too 13:33 sapier sfan5's mod is only half way as it does only support internal strings (at least if I understand correctly what he does there 13:35 sapier imho we need something in between kaeza's internalization and sfan5's localization mod ... maybe the two of them can join forces? 13:35 iqualfragile it think it would be desirable to have a stringtable.lang file 13:36 iqualfragile wich contains a basic pair 13:36 iqualfragile i.e. 13:36 iqualfragile msgmoney: Geld 13:37 kaeza iqualfragile, like this perhaps? https://github.com/VanessaE/homedecor/blob/master/locale/es.txt#L10 13:37 sapier hmm I'd prefere something like: 13:37 sapier table_de_DE['origstring'] = "localizedstring" 13:38 PilzAdam sapier, the logo in the credits tab jumps out of the formspec in fullscreen 13:38 sapier oops :) 13:39 sapier it's not a bug but a feature? 13:39 PilzAdam and the game filter buttons should be always in the bottom left corner too 13:39 sapier that's an issue I can't fix 13:39 sapier at least not without major changes to formspec 13:40 sapier imho this is nothing mission critical and can be fixed later if it's really that important 13:40 PilzAdam well, then place them somewhere where they dont fly in the middle of the screen in fullscreen 13:40 PilzAdam maybe always directly under the formspec 13:41 sapier they are at exactly same position as in window mode 13:41 sapier moving them will break overlays in window mode if this can be accepte I'll do it 13:41 PilzAdam it looks misplaced in any other resolution 13:42 sapier so either break overlay or accept it ... the only possible choices I see atm 14:03 sapier PilzAdam i fixed the moving image ... as well as some other elements not beeing used by now that might have been moving in fixed size mode too 14:24 RealBadAngel one thing left to solve and i think i will release bumpmapping patch 14:25 RealBadAngel http://i.imgur.com/jnH8qDX.png this is 256x sphax texture pack with shaders 14:26 RealBadAngel and this is Haven TP: http://imgur.com/a/v2cwx#0 14:31 sapier looks quite interesting 15:22 hmmmm alright i found the precise commit where the regression happens, 81c863ac4df1530123ef7952713623dadb049bec 15:25 hmmmm i have a feeling that the problem is somewhere around game.cpp:1055 15:26 Jordach hmmmm, which problem is this? 15:26 hmmmm the one where it takes 4-5 seconds to start seeing any blocks load (in multiplayer, at least) 15:26 Jordach yup, that's been my bugbear 15:27 Jordach sometimes, with great enough cpu load, it still happens 15:27 Calinou if someone can fix the 30-second-lag thingy, please do 15:27 Calinou it happens on servers with good hardware, it is a software issue 15:29 PilzAdam anyone objects to merge the mod system change now: https://github.com/PilzAdam/minetest/commit/45fcc9de29de762e8dde20a50db4a3184c830e42 ? 15:35 PilzAdam I guess I merge it then 15:37 Calinou +1 15:37 Calinou less complicated 15:44 PilzAdam is it worth to write something in the forum? or write it in the 0.4.7 release topic? 15:44 Calinou write it in release topic 15:45 Calinou maybe we could use a "News" forum section ;) 15:46 PilzAdam thats actually a not so bad idea 15:46 PilzAdam of course only moderators should be allowed to open new topics in it 15:47 hmmmm how could the person who wrote this not notice such a huge problem 15:48 Calinou you can disable mods per-world 15:48 hmmmm and pilzadam, you as the one who committed it, how could you not notice the regression as well? 15:48 Calinou the functionality is the same 15:48 Calinou or, are you talking about the lag regression 15:48 hmmmm i don't know anything about a "lag regression" 15:48 PilzAdam hmmmm, what exactly is the problem? 15:48 hmmmm i'm talking about the noticable, 4,-5 second delay in between the time you start the game and blocks start appearing 15:48 hmmmm see above 15:48 Calinou alright i found the precise commit where the regression happens, 81c863ac4df1530123ef7952713623dadb049bec 15:49 Calinou ah that 15:49 Calinou it is lag, that's all 15:49 Calinou try typing a command and see 15:49 hmmmm you call that lag? 15:49 Calinou yes 15:49 Calinou when I join servers, typing something like /l takes 5 seconds to reply "invalid command" 15:49 Calinou = 5 second lag 15:49 Calinou servers affected with that bug 15:49 PilzAdam hmmmm, you mean the grey screen after the loading screen? 15:50 hmmmm yes, that's what i mean 15:50 PilzAdam I also had this before, especially on slow servers 15:50 Calinou ^ 15:50 hmmmm that's not the same at all 15:50 hmmmm do you ever play singleplayer? just wondering 15:51 PilzAdam I mostly play singleplayer 15:51 hmmmm how could you not notice a 5 second delay in singleplayer after there had been none 15:51 Calinou I do, but I never notice that bug 15:51 Calinou well, lemme try with latest git 15:51 * PilzAdam reverts the commit 15:51 hmmmm don't bother, i found the exact problem 15:52 Calinou k 15:53 hmmmm ah 15:53 hmmmm you don't notice it if menu clouds are enabled 15:53 hmmmm my bad guys 15:54 hmmmm i guess it wasn't as obvious as i thought 15:55 RealBadAngel looks like i finished the code for bumpmapping 15:55 sapier guys I hate it if you commit changes to files gonna be removed it's a mess to rebase code in this case 15:56 PilzAdam sapier, :-( sorry 15:56 RealBadAngel http://imgur.com/a/1qsCu#0 15:57 Jordach RealBadAngel, much much better 15:57 PilzAdam hmmmm, the grey screen seems to have the same duration with or without menuclouds 15:58 hmmmm i already have it fixed 15:58 hmmmm i know precisely what the problem was and all, and yes, there was a problem despite what you see 15:58 RealBadAngel im going to clean the mess now and prepare a commit 15:58 PilzAdam hmmmm, well, commmit then, Im curious 15:58 * iqualfragile too 16:00 hmmmm compiling 16:01 sapier PilzAdam could you have a try of favorites list with current version? I added string checking for invalid formspec chars but I'm just guessing what's wrong in your case? 16:01 hmmmm what is this 16:02 hmmmm do i need to update minetest_game too? 16:02 PilzAdam yes 16:02 PilzAdam common is dropped 16:02 PilzAdam all mods are moved back to minetest_game 16:03 PilzAdam you now should have only minetest_game and minimal in games/ 16:03 hmmmm and should i have nothing in mods? 16:03 hmmmm for some reason i have the full set of default mods in mods/ 16:03 hmmmm not really sure when that happened 16:04 PilzAdam sapier, completly mess; when I start Minetest, I get a black screen with loading bar then it turns grey, and "C++ exception" in the console 16:05 hmmmm works great 16:05 sapier wow can you post error message? 16:06 PilzAdam sapier, it just says "Error from Lua" and "C++ exception" 16:06 sapier do you have any local modifications? 16:07 sapier I renamed game.* to engine.* in order to avoid name confusion to games 16:07 hmmmm https://github.com/minetest/minetest/commit/f577facf79729a03bb274a27c9f50f0092cd3a40 16:07 sapier that's what happened to me when I had a mixed version 16:07 hmmmm moral of the story is to NOT FREAKING CALL SERVER/CLIENT STEPS WITH A DTIME OF 0 IF TIME HAS ACTUALLY PASSED 16:08 hmmmm anyway. 16:08 sapier do you miss time compression and reversal support hmmmm ? ;-) 16:08 hmmmm pilzadam, did you check out that other commit? 16:08 hmmmm sapier, huh? what do you mean 16:09 PilzAdam "that other commit"? 16:09 hmmmm https://github.com/kwolekr/minetest/commit/db077b603bf015e9292695a381767e65a5bf32b0 16:09 sapier :) "it has to be finished yesterday" 16:09 hmmmm I need someone who gets the warnings to compile that for me and tell me if i need anything else 16:10 PilzAdam will try it 16:10 hmmmm thx. 16:10 hmmmm paste all of the warnings you get 16:13 PilzAdam https://gist.github.com/PilzAdam/5608117 16:14 hmmmm wow you know that's a really pedantic warning setup you've got there 16:14 PilzAdam do you want a log without grep? 16:14 PilzAdam thats the default setup 16:14 hmmmm oh 16:14 hmmmm nah that's okay 16:15 PilzAdam well, the Noise warnings are gone now 16:15 hmmmm it just seemed like there were a lot more warnings than there actually were because you grepped it 16:15 hmmmm all of the dtor warnings should be gone 16:15 hmmmm completely 16:16 PilzAdam its good 16:16 hmmmm alright 16:16 Exio hmmmm: did you try compiling it with clang? 16:16 hmmmm nope 16:16 hmmmm i never really figured out how to correctly switch compilers with cmake 16:17 hmmmm changing the cache contents is pretty hackish and it doesn't work 100% 16:17 hmmmm does it still give warnings with clang or something? tell me before i push this to upstream 16:17 Exio i just did rm -rf build; mkdir build; cd build; CXX=clang++ CC=clang cmake .. 16:17 Exio nope 16:17 Exio but it throws other warnings 16:17 hmmmm what are they? 16:17 Exio what you may want to fix in just one commit 16:17 Exio cadoth.net/~exio4/mt/build.clang.log 16:18 Exio er, add http :P 16:18 PilzAdam hmmmm, changing the cache is not hacky at all 16:18 PilzAdam thats pretty much what you do with cmake -Dsomething=something 16:18 hmmmm when i did that, it failed to compile due to some color undefine thing 16:19 Exio i use a "special" build directory for being able to do rm -rf and start from "scratch" :P 16:19 Exio special directory = ./build 16:21 hmmmm struct class.. 16:21 hmmmm i'll fix all those warnings up in a different commit 16:21 Exio kk 16:21 hmmmm none of them are particularily related to what i just did though 16:22 PilzAdam hmmmm, rename the commit to "Make PilzAdam's compiler happy" 16:22 hmmmm and pushed 16:22 Exio lol 16:23 hmmmm lol 16:24 hmmmm pilzadam, for minetest_game, it'd be nice if we started splitting the files into separate things instead of one huge init.lua 16:24 PilzAdam yea 16:24 hmmmm ideally, init.lua should just be a bunch of include directives... 16:24 Exio is there any reason of why is there a big "mod" for all of that stuff? 16:25 hmmmm ?? 16:25 Exio default seems big 16:25 rubenwardy Compared to a ores mod, a wood mod, a tools mod, a environment mod 16:25 PilzAdam Exio, c55 already said that is wasnt a good decision 16:25 PilzAdam but we cant change it without breaking dependencies 16:26 sfan5 sapier: "sfan5's mod is only half way as it does only support internal strings" what are external strings? 16:26 hmmmm if we were doing this over from scratch, then i guess it would be a good idea to split them up like that 16:26 Exio having compatability can be done with making a virtual-default-mod with aliases and making it have the "splitted" names like rubenwardy said 16:27 Exio very dirty and "hacky" 16:32 Exio wasn't https://github.com/minetest/minetest/pull/727 already merged? 16:32 Exio as for the commit i saw a bit ago for fixing that - at least 16:37 PilzAdam someone really needs to fix run_tests(); I got it failing 8 times in a row 16:38 hmmmm that was on my list of things to do 16:38 hmmmm but let's face it, it's not really that critical 16:42 sapier sfan5 e.g. I need to replace arbitrary text at any given position not only within register functions 16:44 PilzAdam https://github.com/PilzAdam/minetest_game/commits/split 16:44 sfan5 sapier: i'll work on that 16:45 sfan5 CMake Error at /home/stefan/mt-build/build/c55/minetest-minetest-93474c4/build/cmake_install.cmake:63 (FILE): 16:45 sfan5 file INSTALL cannot find 16:45 sfan5 "/home/stefan/mt-build/build/c55/minetest-minetest-93474c4/mods/minetest/mods_here.txt". 16:45 sapier sfan5 I'd support you as I don't want to write an additional i18n for formspec menu I think it'd be just crazy doing so much double work 16:46 PilzAdam sfan5, https://gist.github.com/PilzAdam/5608210 16:47 PilzAdam that should fix it 16:47 sfan5 then push that to minetest/minetest 16:48 PilzAdam done 16:48 sapier could someone plz write a single core modmanager? it's just crap to have all mod information dirstributed throughout many components 16:49 PilzAdam sapier, the only thing a mod manager has to do is place mods in $path_user/mods/ and write load_mode_ = to world.mt 16:50 sapier no I need dependency information to recreate the configure world dialog ... depends on is easy but "is required by" is a lot of parsing work 16:51 sapier and I don't see any core component capable of telling me 16:53 PilzAdam so, is everyone fine with the init.lua split? 16:53 Calinou +1 for splitting the default mod init.lua in many parts, if that's what you're talking about 16:53 Calinou crafting recipes, nodes/items/tools 16:54 PilzAdam Calinou, https://github.com/PilzAdam/minetest_game/commits/split 16:54 Calinou IMO any mod should use a separate file for crafting recipes ._. 16:54 Calinou makes things much cleaner 16:54 Exio better than nothing 16:54 * Exio git pulls minetest and splits it in various mods :P 16:55 Exio _game 16:58 hmmmm hmm 16:58 hmmmm to suppress that dumb "comparison is always true" warning, we'll add -Wno-type-limits 17:02 iqualfragile i have pulled the latest version from git 17:02 iqualfragile put all the lines into world.mt 17:02 iqualfragile moved the mod to basedir/mods 17:02 iqualfragile minetest now tells me 17:02 iqualfragile 18:56:42: ERROR[main]: The following mods could not be found: 17:04 PilzAdam hmmmm, mapgen doesnt leak anymore; just checked with valgrind 17:05 hmmmm great 17:07 hmmmm just wondering, does anybody know what the exact procedure is for changing a CMakeLists.txt? 17:07 hmmmm do you need to delete CMakeCache.txt every time you change something there? 17:10 kahrl my personal voodoo method: if it changes something about external libraries, remove the build directory and restart; if it adds or removes in-project source files just run make 17:11 hmmmm nevermind, it didn't work because my compiler doesn't have it 17:11 hmmmm cc1: error: unrecognized command line option "-Wno-type-limits" 17:11 hmmmm so basically i'm stuck with this until i update to freebsd 10 17:12 kahrl where does that warning come from? 17:12 hmmmm type limit warnings are only activated when -Wextra is specified according to the gcc docs 17:12 hmmmm but i get it all the time 17:13 hmmmm you guys who are using a more recent gcc, do you get -Wtype-limits all the time? 17:13 kahrl I don't see it in PilzAdam's or Exio's logs 17:13 hmmmm well i also saw -Wuninitialized in PilzAdam's logs once 17:13 hmmmm which is also activated with -Wextra 17:14 hmmmm so am i really the only person who sees the constant "comparison is always false due to limited range of data type" warnings? 17:15 kahrl I don't 17:17 iqualfragile nevermind 17:27 sfan5 sapier: L.get added to Localisation 17:28 kahrl When I first start a world it loads all addon mods. It does this until I configure the world (even if I click cancel), then it loads no addon mods. Intended? 17:29 PilzAdam no 17:32 kahrl oh! 45fcc9de29 only changed guiConfigureWorld.cpp but it didn't change the corresponding code in mods.cpp 17:33 PilzAdam just fixing it 17:45 PilzAdam is it ok if I clean the code in server.cpp up that prints missing mods in the same commit? 17:47 RealBadAngel https://github.com/RealBadAngel/minetest 17:47 RealBadAngel if some1 wanna try shaders, code is here 17:48 kahrl PilzAdam: I'd say ok 17:48 RealBadAngel i will upload some texture packs in a few minutes 17:48 PilzAdam I also write all mods that are not world.mt with load_mod = false in world.mt, so its easier to edit the file without the GUI 17:49 kahrl sounds good 17:49 Exio RealBadAngel: what about changing the name of shaders? as a cleanup for "test_shader_X" 17:49 kahrl and shouldn't have any bad side effects with the new mod semantics 17:50 kahrl Any objections to this? Otherwise I'll push it after PilzAdam pushed his fix https://gist.github.com/kahrl/5608391 17:51 RealBadAngel Exio, i can make it later on. i will work with shaders for a long time 17:51 Exio kk 17:53 PilzAdam kahrl, done 17:53 PilzAdam kahrl, what does your patch do? 17:54 kahrl it fixes a problem that Exio's clang noticed 17:54 PilzAdam its good then 17:54 kahrl if you set num_emerge_threads to a value larger than 1 the emerge threads are called "EmergeThread" "mergeThread" "ergeThread" ... and probably segfault at some point 17:54 Exio haha 17:55 Exio that is the bug! 17:55 Exio VanessaE had it in his log 17:55 Jordach her* 17:55 Exio er 17:55 Exio yes 17:55 Exio but don't say anything or i'll get killed! ;P 17:56 PilzAdam Exio, *it 17:56 PilzAdam +s 17:56 Exio wut 17:56 PilzAdam *VanessaE had it in its log 17:56 Exio ah 17:56 PilzAdam "she" is reserved for bots 17:57 Exio x 17:57 Exio er, fails 17:57 Exio i think i'll shutup for now 18:04 hmmmm [01:54 PM] if you set num_emerge_threads to a value larger than 1 the emerge threads are called "EmergeThread" "mergeThread" "ergeThread" ... and probably segfault at some point <----- yeah i know, that's a warning i was going to fix 18:04 RealBadAngel https://github.com/RealBadAngel/sphax256 18:04 hmmmm I guess to fix that i'd have to do something like (std::string("EmergeThread") + id).c_str()? 18:05 RealBadAngel heres texture pack with normal maps 18:05 hmmmm or i could just use sprintf 18:05 PilzAdam hmmmm, see the latest commit in minetest/minetest 18:06 hmmmm does that actually work? 18:06 hmmmm we're both assuming that "" + is being treated as concatenation to a std::string 18:06 * VanessaE pokes PilzAdam with a gender-inappropriate pronoun. 18:07 Jordach munt? 18:07 hmmmm i don't really understand when C++ treats a string literal as a char [] or a std::string 18:07 kahrl literals are always char arrays 18:07 hmmmm then how could + itos(id) possibly work 18:08 kahrl itos returns a std::string 18:08 hmmmm i see 18:08 hmmmm and does log_register_thread take a std::string or a char *? 18:08 hmmmm i honestly forget 18:09 kahrl std::string 18:09 hmmmm this part of C++ is a ghetto, i really wish string handling was consistent 18:09 kahrl const std::string& to be precise 18:09 PilzAdam hmmmm, -> Java 18:09 hmmmm or how about C 18:09 hmmmm if i were doing this project from scratch, i'd choose C 18:14 iqualfragile i would choose python 18:14 Exio i hope that was a joke 18:15 kaeza JavaScript 18:15 Exio HTML5* 18:16 kahrl if I was starting a project like this and I was Notch I'd choose Java 18:18 kahrl is there no block sandbox game written in COBOL yet? 18:18 hmmmm there's a more obscure block game that was started with freebasic 18:18 hmmmm but then they saw the light (or really, the problems) and switched to C++ 18:20 hmmmm it's been in development for a while and they don't have as much progress as say, we do, but they have some nifty features like saving block differences only, servers can be linked together to form a single coherent realm, etc. 18:22 hmmmm anyway i think it'd be really easy to do and we'd save a lot of space, if we save only blocks that had been modified since generation (instead of saving _only_ deltas of blocks) 18:22 hmmmm which would of course break minetestmapper ..but wouldn't be too much of a problem if it were moved into the core like we were discussing one day 18:23 iqualfragile linking servers together would be nice 18:23 hmmmm it's very complicated though 18:23 kahrl hmmmm: is the mapgen fast enough to make that feasible? 18:24 hmmmm kahrl, definitely 18:24 hmmmm map generation is way faster than loading blocks from the db 18:24 hmmmm for me loading a block takes about 30-40ms, whereas generating 5x5x5 blocks takes 100ms 18:25 kahrl but you would still have to make a lookup to see if you need to generate a block 18:25 kahrl not sure how long that lookup would take 18:25 hmmmm less time, of course, but long enough that you'd like to avoid it 18:26 hmmmm i was thinking what if we maintained a bitmap for blocks existing in the db 18:26 hmmmm on initial map load, it's read 18:26 PilzAdam mods like flowers (wich grow) would change many blocks 18:26 hmmmm yeah :/ 18:26 kahrl unless they use DecorationDef 18:26 PilzAdam so just flying over a block would probably modify it 18:26 kahrl oh you mean growing 18:27 hmmmm also i stopped calling it DecorationDef 18:27 hmmmm it's just Decorations 18:27 hmmmm don't really get what the point of appending the def onto things is 18:27 PilzAdam there could be a Lua method to not mark the block as modified when adding a node 18:27 PilzAdam but maybe thats not a good idea 18:27 hmmmm eh, let's work out details when we're actually doing it 18:28 hmmmm right now this is idle chitchat about a feature that's far off 18:28 PilzAdam "Decorations" could be anything, while "DecorationDef" indicates something in the code 18:54 sfan5 ‮hello khonkhortisan 18:55 PilzAdam sfan5, are you sure that this is the correct channel? 18:55 sfan5 no 18:57 PilzAdam any objections: https://github.com/PilzAdam/minetest/commit/bc66f4632372e62be53f3b61cd5cc4a82ee44741 ? 19:01 sfan5 PilzAdam: seems fine 19:34 hmmmm how would you fix the -Woverloaded-virtual on clientobject.h:58/activeobject.h:64? 19:34 hmmmm they're both public in scope, so it's not exactly hidden 19:35 iqualfragile a warning called onderland shoud be added 19:36 hmmmm I think i'll be able to fix this by adding virtual bool getCollisionBox(aabb3f *toset); to ClientActiveObject's definition 19:36 hmmmm but i need someone with clang to check 19:38 hmmmm ah nevermind i was able to do it 19:44 hmmmm why is pos in EnvRef::l_add_item not used? 19:58 hmmmm found a potential bug; environment.cpp:2176 - the 3rd mapnode there is the same as the second, so it never gets the damage of lava from the head 20:01 ShadowNinja Hmmmm: Confirmed. 20:01 hmmmm ...confirmed? 20:01 hmmmm what did you confirm 20:02 ShadowNinja The potential bug. I have noticed it. 20:02 hmmmm oh 20:04 ShadowNinja Is there a plan to merge the scriptapi split soon? And when? 20:07 hmmmm i don't know 20:07 PilzAdam hmmmm, this is not a "potential bug", its already well known; everyone just thought that damage_per_second only applies to the bottom part of the player 20:07 hmmmm people are leaving it up to me to decide it seems 20:07 hmmmm well whatever, i'll fix it 20:07 PilzAdam also checked the fix, works fine 20:07 PilzAdam (i.e. change p2 to p3) 20:08 Exio thexyz / celeron55, (offtopic from here - but), what about making #minetest-delta +if #minetest-dev? ( /msg chanserv set #minetest-delta mlock +if #minetest-dev should do the trick ) 20:13 PilzAdam hmmmm, https://github.com/minetest/minetest/issues/168 is this still up to date? 20:13 hmmmm not valid at all anymore 20:14 PilzAdam closed 20:14 PilzAdam sapier, do all your memory leak fixes also fix this: https://github.com/minetest/minetest/issues/142 ? 20:15 hmmmm obviously it does 20:18 Exio https://github.com/minetest/minetest/issues/177 20:21 Exio hmmm 20:23 sapier PilzAdam yes this was root cause for doing the memory fixes 20:23 hmmmm grr... 20:23 VanessaE so it's my fault? ;) 20:24 hmmmm fix warnings on one compiler, get errors on another compiler 20:24 VanessaE oh wait, wrong one 20:24 Exio haha 20:24 Exio i think https://github.com/minetest/minetest/issues/143 can be closed 20:25 Exio the time passes perfectly for me, with clang 20:26 sapier lol clicking that link crashew conqueror :-) 20:28 hmmmm does anybody have objections to adding -Wno-overloaded-virtual? 20:28 hmmmm (for clang only) 20:33 PilzAdam does this suppress warnings? 20:35 ecube wat? 20:36 ecube sorry, wrong window 20:36 Exio didn't https://github.com/minetest/minetest/issues/218 got "fixed" by sapier very 'recently'? 20:36 PilzAdam its not sure if it was fixed 20:37 PilzAdam it may be a side effect of it 20:38 Exio hmmmm: what about this https://github.com/minetest/minetest/issues/364? 20:39 Exio this can be closed, https://github.com/minetest/minetest/issues/517 i think 20:39 PilzAdam the duplication "outside" of the map still happens 20:40 PilzAdam done 20:40 Exio kk 20:44 Exio https://github.com/minetest/minetest/issues/457 "fixed" 20:47 hmmmm pilzadam, obviously 20:48 hmmmm whenever you have -Wno, that suppresses it 20:48 hmmmm exio, mapgen v6 isn't supposed to have any visible changes at this point, further enhancements to v6 are supposed to go into indev 20:48 Exio ah 20:48 Exio kk 20:48 hmmmm anyway 20:48 PilzAdam I dont know if suppressing warnings is good 20:49 hmmmm i believe minecraft does vary the depth by perlin noise like he says 20:49 hmmmm pilzadam, this one is 20:51 PilzAdam yea, do it if you think its good 20:52 PilzAdam btw, do you want a mingw build log? 20:52 PilzAdam (cross compiling to windows) 20:52 hmmmm not really 20:52 Exio more warnings! 20:58 ShadowNinja Now that common has been removed I think this should be merged https://github.com/minetest/minetest/pull/599 it should allow minetest_mods = common... 20:59 PilzAdam actually the plan was to remove all support for the common system 21:01 ShadowNinja Why? It causes redundancy. 21:01 iqualfragile now that PilzAdam has nothing to do: https://github.com/minetest/minetest/pulls 21:01 iqualfragile ShadowNinja: no, it does not 21:01 iqualfragile the new system is to put all the mods into mods/ 21:01 iqualfragile and then enabe them from the wordls 21:02 Exio so, in case it is going to get "deleted", close that pull request? 21:02 ShadowNinja Ah, so a game is basicaly just game.conf and menu/? 21:03 ShadowNinja Sounds good. 21:03 PilzAdam ShadowNinja, the plan is that all minetest_game similiar games will be just mods 21:03 PilzAdam Ah, so a game is basicaly just game.conf and menu/? <- no 21:03 PilzAdam so all games that would use mods from minetest_game should consider to be mods instead 21:04 PilzAdam e.g. a build mod 21:06 ShadowNinja So you use world.mt to basically build your own game? 21:06 PilzAdam you could see it like this 21:09 Exio https://github.com/ChicoTeam/minetest/commit/0e31a0bae8ba2b4d7e84e9c64710c71bedeee79d 21:15 Exio is this https://github.com/minetest/minetest/issues/177 valid somehow? 21:16 PilzAdam VanessaE, ^ 21:21 VanessaE I don't think that's valid anymore 21:22 VanessaE haven't seen it in a long while. 21:31 ShadowNinja This can be closed https://github.com/minetest/minetest/issues/726 does everyone like the set_node_as idea? 21:34 Exio i wonder 21:35 Exio wouldn't be that https://github.com/minetest/minetest/issues/664 caused by removing the * BS in https://github.com/minetest/minetest/commit/df3c925b3ccae3bdba125e6dc3ecc740739baeab#L4R2068? 22:10 Exio oh, nvm 22:22 PilzAdam anyone wanna test this? https://github.com/PilzAdam/minetest/commits/wieldchange_animation 22:23 PilzAdam ist basically this https://github.com/minetest/minetest/pull/702 rewritten without the issues 22:23 PilzAdam *its 22:24 PilzAdam the only thin that needs to be tested is the animation speed 22:24 PilzAdam +g 22:28 Exio it seems ok for me 22:41 PilzAdam just tested it a bit and it seems to be not annoying 22:48 PilzAdam so, anyone against pushing? 22:49 ShadowNinja Sounds good to me. 23:11 PilzAdam guess I push it hen 23:11 PilzAdam +t 23:28 PilzAdam tested this: https://github.com/minetest/minetest/pull/701 23:28 PilzAdam anyone against it? 23:43 PilzAdam I merge it then