Time Nick Message 00:01 GreenXenith Here to let you know: The 2022 Minetest GAME JAM has BEGUN! The themes for the jam are: SPACE, SECRETS, and STORY. You only need to use one of these in your game, but it is encouraged to use as many as you can (you'll probably score better and have a better game). You can find a countdown timer at https://jam.minetest.net/. Happy jamming! 08:24 definitelya Me when writing and modifying EZ Minetest API calls for a mod, but then stumped when my local variables don't get initialized: '_-_ 08:25 definitelya The Lua API abooser 14:23 mazes_84 I feel less lonely, this week I commited without initialisation -> tonumber(""). :( 18:09 fluxionary_ does anyone know if there's already an issue for the fact that if you press "up" while typing, the client forgets what you were typing? you can press down again, but it's just a blank line. 18:10 fluxionary_ i'm not finding anything, but i'm not sure i'm searching using the right terminology 20:24 giov4[m] do you guys know if entity:remove() can fail for some reason? 20:26 runs Hi 20:26 runs Remember to vote in the SAMMY AWARDS 20:26 runs https://forum.minetest.net/viewtopic.php?f=3&t=28695 20:26 runs vote for your favorite Game, Mod and whatever 20:27 MTDiscord cool but where do we vote 20:27 runs In the post 20:28 runs Look this guy voting: 20:28 runs https://forum.minetest.net/viewtopic.php?p=417194#p417194 20:28 runs and do the same 20:35 giov4[m] "do you guys know if entity:..." <- I also found this error in the console "Pushing ObjectRef to removed/deactivated object, this is probably a bug." 20:35 giov4[m] The entity was attached to a player but when calling remove() it doesn't, well... remove sometimes (I can't reproduce it tho :/) 20:36 sfan5 do you have steps to reproduce such conditions? 20:42 runs first you have to detach, then remove 20:42 giov4[m] runs: but it works most of the time 20:42 runs most of the time is not ALL the time 20:43 runs I do it so, and i am happy, zero errors 20:43 giov4[m] I'll do it then, but it should be written somewhere if that's the case 20:44 runs there are a lot of factors to take into account 20:44 runs it is better to do as I say 20:45 runs my method is the safe way 20:46 MTDiscord removing an entity should probably just automatically detach it if it needs to be detached first. 20:47 MTDiscord It is obvious from the fact that remove() is being called that they cannot intend to have the object remain attached while not existing. 20:49 sfan5 since attachments are involved this sounds like it could be an engine bug, hence I'd like to hear in which conditions this happened 20:51 Krock @Warr1024 the detachment already happens and mods can use attachment callbacks to remove the children 20:51 runs I Normally (90%) they are errors in the mod code, but it could be an engine error. 20:51 MTDiscord If the detachment is already happening automatically then runs' solution shouldn't actually help anything. 20:52 Krock giov4[m]: when you keep track of ObjectRefs outside of callbacks, this can happen very quickly. Don't do that. 20:53 runs Yes, errors are often hidden in the most unsuspected places. 20:54 MTDiscord I mean to be fair, if there were errors in obvious places, then I'd hope that you'd have just fixed them already. 21:00 sfan5 Krock: I suspect that get_attach could return a ref to a removed object 21:04 Krock sfan5: in this case LuaEntitySAO::step would throw a warning 21:04 Krock l_remove() detaches immediately 21:05 giov4[m] "since attachments are involved..." <- I have a skills API that can attach an entity to a player but want to remove it after x seconds; most of the time it works but once in a while it doesn't (but the skill stop, so the function it's called and I've checked that the only condition that could prevent it from working is true) 21:05 giov4[m] So I save the ObjectRef as soon as the skill starts and when it has to be removed I use ObjectRef:remove() 21:05 Krock there's never an invalid ObjectRef because they're all detached before the actual removal (which happens each server step) 21:06 sfan5 hm 21:07 Krock also their mentioned warning log is invalid. it does not exist in Minetest 5.7.0-dev 21:08 Krock nvm. s_base.cpp L462. there's a line break....