Time Nick Message 10:48 sfan5 https://github.com/minetest/minetest/pulls?q=is%3Aopen+is%3Apr+label%3A%22One+approval%22 10:48 sfan5 so who's gonna merge all these? 10:56 Krock I'd do that if they had a second approval 10:57 sfan5 be the approval you want to see 10:57 Krock it's the review & quality control I'd like to see 13:52 clavi can Minetest use PostgreSQL's scram-sha-256 authentication method? 13:52 clavi I tried using it but it didn't work 15:08 Krock will merge #8128 in ~10 minutes 15:08 ShadowBot https://github.com/minetest/minetest/issues/8128 -- Optimize string (mis)handling by osjc 15:09 Krock #8538 too 15:09 ShadowBot https://github.com/minetest/minetest/issues/8538 -- Add content_rating tag to appdata by cpba 15:33 Krock is anyone keen on testing #8385? 15:33 ShadowBot https://github.com/minetest/minetest/issues/8385 -- Add /help formspec for commands and privileges by SmallJoker 16:11 nerzhul clavi i think it depends on the libpq directly 16:12 clavi so it should work? maybe my config is wrong 18:02 VanessaE ServerError: AsyncErr: ServerThread::run Lua: Runtime error from mod '' in callback LuaABM::trigger(): " " is not a registered node! 18:02 VanessaE what causes this? 18:02 VanessaE example crash citing that ^ https://pastebin.ubuntu.com/p/jxrjCg3Sc6/ 18:03 Krock node timer on an unknown node 18:04 Krock oh wait. that's actually core.set_node failing to place a node 18:04 VanessaE er, but citing ABM stuff? 18:05 Krock no, I didn't read it well the first time. Not related to node timers 18:05 VanessaE heh ok 18:06 Krock https://github.com/minetest/minetest/issues/8378 18:06 VanessaE my first thought was set_node() failing on an unknown node and then failing to throw the usual "not placing CONTENT_IGNORE" warning. 18:07 VanessaE AH 18:07 VanessaE leave it to Wuzzy to report it :P 18:08 Wuzzy ??? 18:08 VanessaE nothing, Wuzzy. just smile and nod ;) 18:09 deezl VanessaE, is this related to the crash on BL from basic_machines earlier? 18:09 VanessaE yes. 18:09 VanessaE that IS the crash 18:11 p_gimeno the ghost in #5375 would be me 18:11 ShadowBot https://github.com/minetest/minetest/issues/5375 -- Wheat seeds grow as cotton (wrong ABM seems to be executed on catch-up) 18:12 Krock ikr 18:13 p_gimeno I no longer think it's a compiler bug, it's probably either a race or a problem with using deallocated memory 18:14 Krock I wouldn't expect it to be a race problem since mapblocks are unloaded separately, and node placements only happen rarely 18:15 Krock but wouldn't deallocated memory throw a segfault there? 18:16 p_gimeno no, if you hold a pointer to a block you've allocated and you deallocate it, you can still write to and read from the pointer 18:16 p_gimeno if something else is allocated at the same address, then the pointer is writing into someone else's memory 18:16 VanessaE but shouldn't that trigger a segfault? 18:16 p_gimeno no, deallocated memory keeps being paged 18:16 Krock writing into someone else's memory should trigger that 18:16 Krock if that's allocated 18:17 p_gimeno by someone else's I mean within the same process 18:17 Krock ah 18:17 Krock now, question is why the pointer is not properly nulled 18:17 VanessaE right, can't really trigger a fault then. 18:18 p_gimeno well, it's one hypothesis 18:18 p_gimeno I haven't ever looked into the ABM code 18:19 Krock there's also a cache for the nodes which have triggers - it could also be possible that they're out of sync 18:19 Krock I don't know the ABM code well either, but there are some optimizations to not scan all nodes 18:30 paramat input needed on #8249 18:30 ShadowBot https://github.com/minetest/minetest/issues/8249 -- Make autoforward simulate keys by DS-Minetest 18:47 Krock continuous forward just emulates pressing the forward key. then why is aux1 even involved? 18:48 Krock forward + aux1 is mostly used for sprinting, but that's not like walking forwards normally 19:03 paramat i agree. to get this moving could you support the removal of 'aux' simulation? 19:03 paramat (i mean, add a comment) 19:05 paramat the reason aux is simulated is explained in the long line comment discussion 19:07 paramat because some mods expect 'aux' to be pressed when the player moves fast, however this causes bugs as i explained so can't be done 19:10 paramat i think the author is waiting for a second core dev opinion before they change the PR 19:17 Krock players also fly fast btw 19:21 Krock answer written. 19:22 paramat heh it's a really complex issue 19:22 paramat thanks 19:22 Krock but the implementation is simple 19:22 paramat yes 19:23 Krock no matter how complex the explanation is, it all comes down to whether aux1 is related to autoforward, or not. 19:23 Krock if this cannot be answered, then the PR is superfluous and the old issue will persist 19:25 VanessaE bleh. autoforward. how can that possibly be useful in practice? imho, first, implement cheapie/p_gimeno "fast shift, fast lock" idea and fix that ABM bug. those are more important imho :) 19:26 Krock never used autoforward either 19:26 VanessaE (I mean, useful outside of some kind of flight sim game :P) 19:26 paramat autoforward is extremely useful for me, removes hand stress on long journeys 19:28 VanessaE heh. 19:28 VanessaE define "long" :) 19:33 paramat roughly > 30s 19:33 VanessaE pfft 19:33 VanessaE 30s 19:34 VanessaE you're funny. :) 19:34 VanessaE one of the longer runs on one of my servers takes about 2 minutes. at 3x normal run speed. 19:35 VanessaE (or about 5 minutes at normal speed, including time spent slowing down for areas that'll make you careen off the road into a cliff or whatever) 19:35 VanessaE (the "3x" initiated by a "nitro" station, which uses digilines and mesecons command blocks to temporarily change the player's speed) 19:43 paramat btw autoforward still isn't bound to a key, Krock would you support it being bound? maybe F? 19:43 VanessaE oh HELL no. 19:43 VanessaE that's too easy to hit during normal play 19:43 VanessaE pick something on the right side of the keyboard. 19:44 VanessaE maybe N or something. 19:44 VanessaE (it's not like JKL are obvious either, for their respective functions) 19:51 paramat logically it should be close to the movement keys just like 'move fast' and 'sneak' were. concern about accidental presses just leads to controls not being where they should be, players just have to learn 19:55 paramat but sure, any key is better than none 19:58 paramat maybe i've forgotten but, why do we have key bindings for 32 hotbar slots when there aren't enough spare keys for 32? 19:59 paramat 16 seems about right 20:00 VanessaE because I thought it logical to do so. 20:00 Krock paramat: no, I think it should be kept empty. Players may define it when they want to use it 20:01 VanessaE plus, future-proofing, someone out there will have a gaming keyboard with a couple dozen extra keys, or maybe one day, 10 years from now :P Minetest will support modifiers such that maybe Ctrl-1 could be slot 11, Super-1 is 21, or whatever. 20:02 paramat actually, 24 20:03 Krock VanessaE: those keyboards will be hard to use with only two hands 20:04 VanessaE Krock: maybe players will evolve three hands? :P 20:04 paramat almost no-one has keyboards with lots of extra keys 20:04 VanessaE "almost". 20:04 VanessaE granted, I don't. 20:05 VanessaE but someone will. 20:05 paramat and almost no-one needs instant key access to 32 slots 20:05 Krock well, you could still make use of the numpad 20:05 VanessaE plus, keypad. doesn't even work right at the moment, so there's 16 keys players could binfd 20:05 VanessaE damn it, ninja'd 20:05 Krock hah 20:07 paramat remembering which key is which slot is only really possible with number keys + 10 keypad keys, that's still less than 24 20:07 VanessaE and back to modifiers, think Ctrl, Super, Alt with the F1-12 keys as well as the top row number keys/ 20:08 VanessaE why 24? I thought I defined all 32? 20:08 VanessaE *looks* 20:08 paramat i mean, my suggestion was 24 =) 20:08 VanessaE yep, I did. 20:08 VanessaE nah, don't reduce if there's no practical reason to do so. 20:09 paramat anyway nevermind 20:09 VanessaE it's like it takes more than a couple dozen bytes of memory for those last 8 bindings :P 20:09 VanessaE it's not like* 20:11 paramat merging #8481 thanks for review 20:11 ShadowBot https://github.com/minetest/minetest/issues/8481 -- Allow multiple cave liquids in a biome definition by paramat 20:12 Krock np, thanks for the patience. I didn't have the chance to test PRs earlier 20:18 paramat that's ok, i know you're busy 20:18 paramat so merging game#2358 too 20:18 ShadowBot https://github.com/minetest/minetest_game/issues/2358 -- Make cave liquids biome-determined by paramat 20:18 Krock +1 20:20 paramat thanks 20:24 paramat good, cave liquids dehardcoded, dungeons next 20:51 rdococ would it be okay if the callbacks weren't saved across server restarts? 21:02 rdococ there might be a way to fix it yet 21:02 rdococ nvm