Time Nick Message 09:29 erle MisterE123 i got no reply to https://github.com/minetest/blog/issues/129 by you so far, but you are assigned. anything i should do? 10:47 jonadab Hmm. My instinct on seeing that question would've been to point to the /emergeblocks documentation. Technically not what was asked, but I'm used to seeing XY problems, and that's where my brain would've gone. 10:47 jonadab (The question about generating empty blocks, I mean.) 11:44 MTDiscord Looks like it will go in to the blog. 11:50 [MTMatrix] create ghost object for build - this mod exists? 13:39 MinetestBot 02[git] 04sfan5 -> 03minetest/minetest: Fix double-free of minimap textures 132479d51 https://github.com/minetest/minetest/commit/2479d51cc665f8b29b919ad3158444d032840b82 (152023-09-13T13:27:07Z) 20:04 elelelelel2010 h 20:04 elelelelel2010 hi 20:27 [MTMatrix] How i can do "sleep" (wait some seconds) in function? 20:28 [MTMatrix] How i can do "sleep" (wait some seconds) inside function? 20:28 [MTMatrix] Just for slowdown on_place speed 20:29 calcul0n use minetest.after() 20:29 [MTMatrix] wait, but if only for call outside function? 20:29 [MTMatrix] wait, but it only for call outside function? 20:29 calcul0n a real sleep would block the whole engine 20:30 calcul0n you'll have to write things differently but it's not very hard 20:31 [MTMatrix] eg. minetest.after(5, func), but in my case if i do on_place = minetest.after(5, func), it causes error 20:32 calcul0n yep, that would only call after at registration time 20:32 calcul0n you need on_place = function(...) minetest.after(5, func) end 20:33 [MTMatrix] oh.. thanks 20:35 MTDiscord erle: we made a little discovery that since [png blindly passes on the image data to irrlicht as an in-memory file, it works for inlining images of any format, not just PNG 20:35 erle ROllerozxa this makes the grammar even stupider, but also this is useful lol 20:36 MTDiscord sfan5: fixing that is trivial. Just replace __temp_png with __temp.png. 20:37 MTDiscord (no, it won't, irrlicht doesn't care about file extensions, only the header) 20:37 erle ROllerozxa i guess that means you can pretty much start using unicode_text on signs without doing the dynamic media dance 20:37 MTDiscord yep 20:37 erle i mean, it's super fucking ugly (i actually cautioned against adding anything like [png or [tga or so), but it's obviously useful. so have fun with it! 20:38 MTDiscord It doesn't care only because __temp_png doesn't have an extension. 20:39 erle savilli are you playing “let's break some useful feature because itwas not intended” or what? 20:39 MTDiscord the fun police has arrived, thankfully they can't stop us 20:39 erle FUN is banned! 20:39 erle btw, i actually think the dynamic media API is pretty stupid for stuff like signs and so on, for all the racy conditions. 20:40 erle it would be much better if this would work more like a web browser, where you can reference a texture and the client loads it later. 20:40 MTDiscord I'm just eating my popcorn. You guys are very entertaining with that tga discussion. 20:40 erle at least for signs 20:40 erle savilli i am amused too. i mean the whole minetest.encode_png() and [png was kinda not too well-thought-out in the first place. 20:41 erle this is probably the funniest bug i have seen in a while 20:41 erle i think since the object oriented enchanted book probably 20:41 erle the bug was like this: books needed to be enchantable in mcl2 to have an enchantment. so … you could use a stack of books. and enchant the stack using one book. 20:41 erle because nothing prevented you from enchanting stacks of items 20:41 celeron55 obviously the "png" in the texture modifier doesn't refer to the image format but actually means Potentially Not a Graphic 20:42 erle and weapons are not stackable in mcl2 20:42 erle IF ONLY someone had warned the implementors and said “you need to have a grammar for texture modifiers that is context-free or regular” 20:42 erle oh wait, i did that 20:42 erle and i think luatic did too? 20:43 MTDiscord I thought it means Pleasure Not Guaranteed 20:43 erle anyway, this is the most monkey's-paw thing i have seen so far and evidently useful to implement a signs mod. 20:43 erle because so far the problem is that you need lots of bookkeeping for the players and the textures 20:44 erle with this way however, you can just replace whatever generates your texmods with unicode_text and call it a day. well, i need to fix 1 bug in the bidi algorithm (or disable it) i guess. you'll find it soon enough (uniess i get there first). 20:45 erle yeah so 20:45 erle uh 20:45 erle LANGSEC told you so! :P 21:01 [MTMatrix] hm, so strange, it still crashes server... invalid invocation 21:20 erle ROllerozxa i have learned from my question “can i change the main menu script from a mod” that FUN is best kept to oneself to not invite the FUN POLICE :P 21:25 erle by which i do mean of course that i have no idea how to achieve code execution via a crafted ogg vorbis file 21:27 erle but readers of CVE-2020-20412 might be able to figure that out! 21:28 erle (i am not good with computers) 21:29 [MTMatrix] so... the 'while true do if math.random(1,10000) < 5 then break end end' very bad idea for use? 21:29 erle for what use 21:29 erle heating your room, it might be useful 21:30 erle depending on speed of device obv 21:30 [MTMatrix] ahahah 21:30 erle and bias of RNG 21:32 erle localhost can i ask you a question 21:32 erle or maybe two 21:33 [MTMatrix] maybe, but no warranty answer 21:33 erle okay, are you on windows? 21:33 [MTMatrix] no, linux :D 21:33 erle which one? 21:35 [MTMatrix] Server side or client side? (but of course I can't imagine Windows server running minetest) 👀 21:35 erle client side 21:36 [MTMatrix] Okay, void 21:36 erle what's your version of libvorbis 21:36 erle i mean the one you compiled minetest with 21:36 [MTMatrix] Wait a sec 21:37 [MTMatrix] libvorbis-devel-1.3.7_1 21:37 [MTMatrix] hack via sound? 👀 21:37 erle that's too new for CVE-2020-20412 21:37 erle https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2020-20412 21:38 erle > lib/codebook.c in libvorbis before 1.3.6, as used in StepMania 5.0.12 and other products, has insufficient array bounds checking via a crafted OGG file 21:40 erle but vorbis_version=1.3.7 21:40 erle i am disappoint! 21:42 [MTMatrix] eh... debian stable... so stable CVE 21:46 erle localhost this list is too short, you can help by expanding it! https://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=minetest 21:52 [MTMatrix] fly speed violation is CVE? 22:57 erle localhost i doubt it