Time Nick Message 02:21 ANAND I think #7336 is a very simple PR. It's got one approval already - could another dev or two be able to review it if possible? Thanks in advance :) 02:21 ShadowBot https://github.com/minetest/minetest/issues/7336 -- Implemented hasprivs chat-command (Closes #7266) by ClobberXD 02:23 Shara If I could, I would, because I really would like that added. :) 02:26 ANAND CLANG_TIDY fails to build, though. Considering that I did nothing to the C++ part of the engine, that's quite strange 02:34 paramat the clang tidy check itself breaks sometimes, code may be fine 02:34 ANAND Ah, I see... 02:35 paramat you can usually tell by reading the clang tidy report itself 02:37 ANAND I tried to understand what the report says, couldn't get my head around so much jargon :) 02:38 * ANAND is a noob, if you haven't realized that already ;) 02:38 paramat i can't either, but if it says "blah failed to load" "can't find blah" then it's broken 02:39 ANAND XD 02:39 paramat if it's working it will be obvious, it will specify code improvements 03:07 paramat merging game#2139 in 5 mins 03:07 ShadowBot https://github.com/minetest/minetest_game/issues/2139 -- Spawn mod: Also reposition players on respawn by paramat 03:14 paramat merging 03:15 paramat merged 06:38 nerzhul ANAND, yeah i should add like some other checks verification the diff modified the C++ part 09:28 ANAND Oh ok :) 09:35 ANAND It's just that I think this should be in 0.5.0 itself. Sorry if I come across as pushy :) 09:35 ANAND Especially given that 0.5.0's feature freeze is right around the corner 10:21 nerzhul we are not in feature freeze, but we should talk about it. One feature is missing about CSM before the feature freeze 11:18 ANAND None of the packages show up when I click on "Browse online content" - all I get is an empty screen. I'm on 0.5.0-dev-75aa41c6 btw 11:19 ANAND Am I making a silly mistake, or should I make an issue out of this? :P 12:48 ANAND Had this question for quite a long time: Is multi-threading capabilities going to be added to Minetest anytime soon? 12:48 ANAND Would it be reasonably easy to implement? 13:50 Fixer "Apps built using OpenGL and OpenCL will continue to run in macOS 10.14, but these legacy technologies are deprecated in macOS 10.14. Games and graphics-intensive apps that use OpenGL should now adopt Metal." 13:50 Fixer Bye-bye minetest on iCrap 13:53 Krock Never heard of that metal stuff 13:58 Kray metal, vulkan and d3d12 are practically the same thing 14:09 VanessaE these "nil" glitches that people say are happening with 0.4.17... is this related? https://gitlab.com/VanessaE/homedecor_modpack/issues/2 14:11 Krock it's exactly what xerox123 (?) reported yesterday 14:11 VanessaE oh ok 14:11 VanessaE would you mind weighing in on it? 14:11 Fixer irrlicht does not support dx12, vulkan or metal 14:13 VanessaE Krock: oh, well that's right, he reported the homedecor bug over in #-hub, but I meant the part where I can't reproduce it 14:13 Krock This error is caused by bugfixing this error #7044 14:13 ShadowBot https://github.com/minetest/minetest/issues/7044 -- MetaDataRef:set_int does not check argument types 14:15 Krock however, changing it to a valid :set_string("", "") call will result in the same effect 14:17 VanessaE https://gitlab.com/VanessaE/homedecor_modpack/blob/master/homedecor/handlers/water_particles.lua#L112 14:17 VanessaE hm 14:18 Krock VanessaE, http://irc.minetest.net/minetest-hub/2018-06-06#i_5324303 14:19 VanessaE right, but setting a key to the empty string would still leave a key reference though, but nil'ing it would delete the key entirely. 14:19 VanessaE or at least that's how Lua usually operates. 14:20 Krock but it's not how the meta storage works 14:20 VanessaE ah 14:20 Krock Code reference for that empty strings will be removed (see also lua_api) https://github.com/minetest/minetest/blob/stable-0.4/src/metadata.cpp#L88 14:22 VanessaE gotchya. 14:23 VanessaE thanks. fixed. 14:24 Krock yw. nice :) 14:28 VanessaE incidentally: gitlab autocloses if you write "fixes #xxx" in a commit message, similar to the other site. 15:09 nerzhul github too 16:46 rubenwardy A commit introducing a blocker issue was added to 0.4.17 16:47 rubenwardy #6798 16:47 ShadowBot https://github.com/minetest/minetest/issues/6798 -- Fatal error in void log_deprecated() when 'deprecated_lua_api_handling = log or error' 16:48 Krock and #7419 is the dupe of that ^ 16:48 ShadowBot https://github.com/minetest/minetest/issues/7419 -- 0.4.17 server crash 16:50 rubenwardy Lololol 16:50 rubenwardy Sorry 16:50 Krock again? like what 16:50 Krock but previously we managed to close it twice 16:51 rubenwardy I wasn't aware that this commit was in 0?4.17 16:51 rubenwardy Hence why I didn't prioritise #6708 16:51 ShadowBot https://github.com/minetest/minetest/issues/6708 -- Add file and line to NodeDef field deprecation warning by rubenwardy 16:56 Krock rather than working with offsets, it would be best to provide an entire backtrace ^ 16:56 Krock since we don't know whether the deprecated log call was triggered by a called builtin function or directly from a mod 16:57 Krock related PR #7377 16:57 ShadowBot https://github.com/minetest/minetest/issues/7377 -- minetest.log: Add mod origin by HybridDog 17:01 rubenwardy Yeah 17:01 rubenwardy Might just revert sfan5's commit 17:01 rubenwardy Thanks Megaf 17:01 rubenwardy ;) 17:01 sfan5 uh 17:02 Krock that's not how you solve that problem 17:02 sfan5 surely cherry-picking the fix is easier 17:02 rubenwardy It is how you solve it 17:02 rubenwardy There's already a backtrace 17:02 rubenwardy It's behind a setting 17:02 rubenwardy Sfan's commit added the offset thinf 17:02 rubenwardy Mine tried to correct it to the right file and remove the crash 17:03 rubenwardy But that's hard without pointing at builtin 17:04 Krock > It's behind a setting , do you mean "deprecated_lua_api_handling"? If so, then it's not a proper backtrace 17:04 rubenwardy There are 3 options: my pr with built-in knowledge, my pr without pr knowledge (may blame builtin for mod issues), revert bug introducing commit 17:05 rubenwardy Oh, it's to infostream 17:05 rubenwardy So the option will be log level 17:05 rubenwardy Also, option 4: revert and promote to warning or whatever 17:08 rubenwardy My point is to do "rather than working with offsets, it would be best to provide an entire backtrace" you'd revert the commit 19:10 Krock rubenwardy, would you mind updating #6708 with the requested changes plus rebasing? so we can check whether it returns the correct line in each case? 19:10 ShadowBot https://github.com/minetest/minetest/issues/6708 -- Add file and line to NodeDef field deprecation warning by rubenwardy 19:44 rubenwardy Krock: it won't 19:44 rubenwardy see my comment 19:44 Krock the "let's revert the other commit" comment? 19:45 rubenwardy "What should be done here? This PR fixes a crash when calling a deprecated field, however the location it gives may not be accurate. 19:45 rubenwardy I'm not sure how to properly pass in the path to builtin to work out where in the stack the last entry before builtin. I could instead try and find builtin in the path, but this will break if any mods or directories in mods are called builtin. 19:45 rubenwardy This could be merged as-is, as it's still more accurate than the current code on master and doesn't crash" 19:46 Krock ah, within the PR. I see. 21:18 paramat reverting https://github.com/minetest/minetest/commit/813d819d153c78eb81893c208ba73c56237ea3d4 in 0.4.17 seems ok to me, that doesn't need to be backported to 0.4.17 21:35 paramat i had a feeling 0.4.17 wasn't tested enough, no build was available to increase testing :) 23:30 paramat will merge #7422 in 15 mins 23:30 ShadowBot https://github.com/minetest/minetest/issues/7422 -- Lua_api.txt: Warn of errors possible with use of VoxelArea:index()/indexp() by paramat 23:45 paramat merging 23:47 paramat merged