Time Nick Message 17:18 rubenwardy merging #12484 in 10 17:18 ShadowBot https://github.com/minetest/minetest/issues/12484 -- Fix crash when `item` isn't provided to core.hud_replace_builtin by rubenwardy 17:28 Krock oh shoot I was commenting on that 17:28 Krock should be core.PLAYER_MAX_BREATH_DEFAULT * 2 17:28 Krock rubenwardy: ^ 17:29 Krock well I guess it doesn't matter too much since it's just a fallback 17:29 rubenwardy you should probably say in IRC first (: 17:29 Krock well yeah I was testing it without checking IRC 17:29 rubenwardy why *2? Aren't those values already in hp etc 17:30 Krock statbars use halves 17:30 Krock 10 * 1/2 = 5 bubbles 17:30 rubenwardy so does hp 17:30 rubenwardy is breathe different to hp? 17:30 Krock 20 * 1/2 = 10 hearts for HP. yet HP and Breath have different scaling 17:30 rubenwardy that's dumb 17:30 Krock can confirm 17:31 rubenwardy well, this is used as nominal value so I think it'll be scaled to 10 hearts automatically 17:31 Krock "item" is the nominal for this statbar 17:32 Krock > Other than that, the HUD documentation sucks (unclear purpose of fields) 17:32 Krock from my yet-not-sent review comment 17:33 Krock whatever. let's wait for the next bug report 17:59 erle rubenwardy, i think crash fixes (like this thing) sholud have unit tests, but i do not know how to make a test that does not crash when the code runs into a lua crash. is this possible with the current minetest setup? 18:01 rubenwardy pcall 18:03 MTDiscord busted has assert.has_error for this, which basically does the pcall for you 18:46 rubenwardy merging #11663, #12307, #12495, #12496 in 10 18:46 ShadowBot https://github.com/minetest/minetest/issues/11663 -- Allow to set maximum star opacity at daytime by Wuzzy2 18:46 ShadowBot https://github.com/minetest/minetest/issues/12307 -- DevTest: Add item meta editor by Wuzzy2 18:46 ShadowBot https://github.com/minetest/minetest/issues/12495 -- Optimize strict.lua by appgurueu 18:46 ShadowBot https://github.com/minetest/minetest/issues/12496 -- Serialization spec: Fix number fuzzing by appgurueu 19:13 rubenwardy merged all except #12496 19:13 ShadowBot https://github.com/minetest/minetest/issues/12496 -- Serialization spec: Fix number fuzzing by appgurueu 19:42 rubenwardy Krock: I hope you don't mind updating settingtypes.txt in #12450, it would be good to have the GUI match the C++ limits 19:42 ShadowBot https://github.com/minetest/minetest/issues/12450 -- Enforce limits of settings that could cause buggy behaviour by SmallJoker 19:43 Krock yes I'll work on it sooner or later 19:43 Krock I don't agree with the chunksize limits, hence will update settingtypes.txt instead 19:43 rubenwardy that's fair 19:58 rubenwardy merging #12496 19:58 ShadowBot https://github.com/minetest/minetest/issues/12496 -- Serialization spec: Fix number fuzzing by appgurueu 20:03 Krock was there anything wrong last time? 20:04 rubenwardy huh? 20:04 rubenwardy last PR? 20:04 rubenwardy the problem is that `math.random` only supports upper with max 2^32 20:04 rubenwardy and it's passing in 2^52 20:08 Krock > merged all except #12496 20:08 ShadowBot https://github.com/minetest/minetest/issues/12496 -- Serialization spec: Fix number fuzzing by appgurueu 20:09 rubenwardy I wanted to look into the Lua code 20:09 rubenwardy like, the .c code 20:09 rubenwardy to understand how it works 20:09 Krock ah. Lua interpreter source. yes. makes sense. 20:09 rubenwardy which is why I delayed 20:09 rubenwardy I should have probably started from 10 again? 20:10 Krock ? guess it does not matter 20:10 Krock after all it's testing code, and if something's wrong, it should tell that by itself 20:10 rubenwardy I feel like fixing a bug. Shame Minetest doesn't have any 20:10 Krock are you kidding? 20:12 Krock what would you think of digging deeper into the Android error in #7629 ? :3 20:12 ShadowBot https://github.com/minetest/minetest/issues/7629 -- Run Minetest update checker on startup by SmallJoker 20:14 rubenwardy god I hate Android NDK 20:14 Krock I take that as a yes 20:31 erle Krock rubenwardy what is it about the chunksize limits in https://github.com/minetest/minetest/pull/12450/files ? 20:32 Krock you're not supposed to change it anyway 20:32 erle i am pretty sure mineclone5 changes it to 4 20:32 erle but i would have to look 20:32 Krock but if you - for whatever reason - need to change it, it's still allowed to have a certain reasonable range 20:33 erle i gather that chunksize is limited to 10 now 21:07 erle i can't find the compilation instructions for minetest with OGLES2 and the framerate on reform2 is worse-than-on-a-15-year-old-thinkpad with desktop GL. could anyone help me there? 21:24 MTDiscord uncommenting a define in irrcompileconfig.h to enable GLES2 support in irrlichtmt and then compile minetest with ENABLE_GLES=ON as a cmake option 21:33 erle ROllerozxa why is the uncommenting a define thing needed? can the cmake option not handle it? 21:35 rubenwardy because irrlicht 21:38 MTDiscord the cmake option handles it for minetest but not for irrlichtmt, irrcompileconfig.h is mostly a relic from the old makefile system upstream irrlicht uses to customise what gets compiled in 21:40 MTDiscord (I say mostly because there's some portions that are used to compile with proper defines on each platform... those could stay but rest should honestly be redone as cmake options for easier configuration) 21:57 Zughy[m] 4 core dev PRs with one approval ? we could reach 80 PRs 21:57 Zughy[m] which I think it's a new record? 21:59 MTDiscord :eyesup: 21:59 rubenwardy well, technically the record is 0 from sometime in 2012 22:05 Zughy[m] then let's close them all and we're gonna reach that record again 22:07 erle i think with that approach Zughy[m] is ready for the highest levels of management ;) 22:07 erle ROllerozxa can cmake alter the defines maybe? 23:09 Pexin III'm confused about #12501, is this a problem with the user's graphic drivers, or what? clouds are supposed to be translucent.. also, should that be "roadmap:needs approval"? 23:09 ShadowBot https://github.com/minetest/minetest/issues/12501 -- Cloud layer transitions are abrupt and too late 23:09 Pexin uh. in 12502 rather 23:10 schwarzwald[m] erle: If you make a PR to handle the OpenGLES setup entirely from CMake I'll review it. 23:24 Zughy[m] Pexin: sounded like a graphical improvement 23:35 Pexin Zughy[m]: well I can stand above a cloud and see the terrain through it. seems to me the user is having issues with graphic driver or possibly gamma.