Time Nick Message 00:02 paramat fixed item entity motion on slippery nodes #6300 00:02 ShadowBot https://github.com/minetest/minetest/issues/6300 -- Item entity: Fix motion on slippery nodes by paramat 01:35 Natechip guys. seriously these bots are taking over the servers.... we need to be able to stop them somehow? Theres more bots than players now! 01:36 Natechip normally they have a name with three letters at the end, some are programmed to move around so they dont get afk kicked... also some kill there selfs on purpose so they dont get afk kicked. 01:37 bigfoot547 Just kick them 01:38 Natechip bigfoot547: think I have all day to kick bots which BTW have autojoin... so if u kick one another one comes 01:38 bigfoot547 Okay 01:38 Natechip you can ban them either. there too many IPs 01:38 bigfoot547 like a hydra :P 01:38 Natechip you can ban them either. there too many IPs 01:38 bigfoot547 Ban the range? 01:38 Natechip constantly changing 01:38 Natechip cant 01:38 Natechip tried. 01:38 bigfoot547 Hmm 01:39 Natechip I would like to not use a three # at the end of a name block either. it wont let ppl with names like yours on. 07:08 nore #5274 is done; anyone to review please? :) 07:08 ShadowBot https://github.com/minetest/minetest/issues/5274 -- [WIP] Clientside translate by Ekdohibs 07:33 nerzhul done 07:37 nore nerzhul: what do you mean by "std::wstring should be a reference instead of a return"? 07:38 nerzhul don't do a copy return, instead pass a parameter as pointer or reference and write on it 07:42 nore hmm, ok 07:48 nore nerzhul: changes done 08:06 celeron55 on windows, wstring is 16-bit, which will still break unicode characters that are multi-byte in utf-16 when processed in this way, right? 08:06 celeron55 on linux it's 32 bit and might work for all characters 08:09 nore celeron55: there is use of wstring, but it always comes from utf8, so it should cause any problems 08:09 nore nothing is sent as wstring 08:09 celeron55 yeah, the protocols are fine 08:10 nore so where is the problem, do you think, then? 08:10 celeron55 that's why i'm not really worried 08:10 nore I'm not sure I understand what you mean 08:11 celeron55 meh, nevermind, and those characters i mentioned are probably broken somewhere else anyway 08:11 celeron55 and, maybe this code works regarding to those anyway 08:12 nerzhul you can send wstring over protocol 08:12 celeron55 no, human language strings should always be serialized as utf-8 08:13 nerzhul oki 08:13 celeron55 it's the way that causes least problems when dealing with multiple platforms and whatnot 08:18 nerzhul right 11:31 nerzhul nore, https://travis-ci.org/minetest/minetest/jobs/267082757 11:31 nerzhul please look at LINT build for all codestyle problem 11:31 nerzhul if some LINT results are not good it's possible to add markers to ignore sections 11:32 nerzhul or the whole file (i don't recommend) 11:40 nerzhul and if you want to be lazy to format files, ensure you have clang 3.9 or greater , place yourself are the project root and type clang-format -i to reformat directly the file 11:40 nerzhul (yeah we should have clang-format or the other shadowninja presented item in MT, code style should just be handled by a tool :p) 14:15 nore ~tell nerzhul the build seems to be failing because clang seems to want me to align some code with a tab + spaces instead of only spaces, what should I do? 14:15 ShadowBot nore: O.K. 14:18 celeron55 wait, what are you indenting and how does it want you to do it? 14:21 nore celeron55: I have something like [tab][tab]errorstream << ... and on the next line, [tab][tab][spaces to align]<< 14:21 nore and it wants me to do the second line with [tab][tab][tab][spaces], which assumes tabs are 8 wide 14:22 celeron55 just saw that 14:22 celeron55 it's clearly wrong 14:22 celeron55 either two tabs always or spaces only after the basic tab indent level 14:22 celeron55 those would be correct in my opinion 14:23 nore well, what surprises me most is that it doesn't complain about errors elsewhere 14:24 celeron55 either errors are disabled elsewhere or someone has corrected the indentation to a broken one elsewhere 14:24 nore hmm yeah, maybe 14:26 nerzhul yeah it's wrong, ignore it for the spaces 14:27 nerzhul celeron55, clang-format have a file whitelist, other files are correctly formated 14:27 nore nerzhul: so, what should I do to fix the build then? 14:28 nerzhul it's the cpp file ? 14:29 nerzhul you have two choice, either add it to clang-format-whitelist or mark the section with // clang-format off & // clang-format on 14:29 nerzhul i suggest first 14:29 nore nerzhul: yes 14:29 nerzhul we don't statuate on the second method 14:29 nore hmmm how do I add it to the whitelist? 14:32 nore ah found the file 14:46 celeron55 why are we using a formatting tool that is unable to format reasonably? 14:46 celeron55 this is nonsense 14:48 Megaf celeron55, which tool? 14:49 celeron55 at least having to add every file that is using stream operators to a whitelist is nonsense; is there other nonsense? 14:51 celeron55 is there not a formatter that's like a python script or something that can actually be customized for a real use case like this? 15:06 sfan5 Megaf: clang-tidy 15:06 Megaf ah 15:06 Megaf not a good idea 15:17 celeron55 the issue isn't really the concept of code formatting, but it seems every tool for that purpose is crap 15:21 nerzhul clang-format 15:21 nerzhul clang-tidy is for the code quality 15:21 Megaf doesnt gcc have a formating tool? 15:22 nerzhul no, only clang provides such tools 15:22 nerzhul (as a compiler suite, i mean, there is astyle presented by ShadowNinja) 15:22 Megaf did you try AStyle? 15:22 nerzhul not myself, ShadowNinja did 15:23 nerzhul but yes finding the correct tool for code format is required, it's not interesting to review code style 15:23 nerzhul clang-format has some cons, but works on ~30 of our code (file based count) 15:23 Megaf well, if don't want to have too much manual work configuring the tool I think the options are either clang-format or AStyle 15:27 nerzhul both tool should be configured 15:31 Megaf Yep, now see Uncrustify, that's way worse to set up 15:31 Megaf there's an interface for it tho, Universal Indent GUI 16:20 nore oh btw, I'm asking again my question from yesterday: is there a simple way I can translate LANG_CODE in all gettext files, or do we have to go through weblate? 16:37 sfan5 why not just use the "language" setting? 16:37 sfan5 or is that not populated if autodetected? 16:38 nore it's not populated if autodetected, unfortunately 16:38 nore it's only there to force the language 16:39 nore and there seems to be no reliable way to detect the language or to ask gettext to give it 16:39 sfan5 that's stupid 16:40 nore yes 16:40 nore (besides, we want languages such as fr and fr_FR.utf8 or things like that to be detected as the same language) 16:43 sfan5 well you can take $LANG and cut off after the first _ 16:43 sfan5 but that only work on linux 16:44 nore yep 16:44 sfan5 just asking gettext for it's detected lang would be ideal 16:44 nore and there is LANG and LANGUAGE that are both used as well 16:44 nore (also, cutting LANG wouldn't work: we have zh_CN and zh_TW) 16:44 sfan5 (LC_ALL and stuff also exist) 16:45 nore (indeed) 16:45 paramat #6256 tested and +1 can anyone review? 16:45 ShadowBot https://github.com/minetest/minetest/issues/6256 -- Alternative code for slipping by bendeutsch 17:44 paramat ~tell Krock could you look at this bugfix? https://github.com/minetest/minetest/pull/6241 since you commented on the issue, fix seems correct 17:44 ShadowBot paramat: O.K. 17:44 paramat +1 for that 19:26 paramat #6241 is fairly obvious and trivial actually, labelled it as trivial 19:26 ShadowBot https://github.com/minetest/minetest/issues/6241 -- Do not add base position to player selection box by stujones11 22:36 KaadmY BTW "Infinite view range" is still incorrect 22:36 KaadmY Since it's not range, but instead if everything is drawn 23:24 paramat yeah 23:27 paramat hmm which issue/PR is this? 23:35 paramat 'view all loaded world' might as well be precise :] 23:56 VanessaE how about calling it "Unlimited view range"