Time Nick Message 03:18 MTDiscord https://cdn.discordapp.com/attachments/747163566800633906/974148534586708009/Screenshot_from_2022-05-11_20-16-17.png 03:18 MTDiscord a new kind of dark mode? 03:19 MTDiscord is this likely a new bug (built from latest master) or did I manage to mess something up in the config? has anyone tested it lately? 03:32 MTDiscord I think that that is just what happens when you try to emulate minetest on android. You have to run it on a device 03:39 MTDiscord is there an issue for this? it seems like something that should be fixable 04:42 MTDiscord I've seen something similar with gui_scaling I think 05:16 MTDiscord heh, super secret android dark mode setting :P 05:35 paradust I believe I know the reason, still trying to fix it 06:17 paradust negative, apparently not 06:18 paradust only thing I've determined for sure is that Android Studio is garbage 06:39 Pexin I tried to figure out android studio 3 years ago and it did not work for me. could not find any alternatives (did not know where to look). I do have _slightly_ better hardware now, would appreciate advice. (maybe not in this channel, I dunno) 06:50 paradust Pexin: you had the same problem? 06:51 paradust I've determined something is going wrong with the OpenGL ES 2 emulation. There's several different modes available, and they change the distortion in different ways, but it never looks normal. 06:52 paradust There's a complete passthrough but it crashes the emulator before it can do anything. 06:53 paradust Even the full software emulation (SwiftShader) has borked output 06:53 paradust it's possible minetest is doing something to trigger this bad behavior, but it'll take a while to figure out 06:59 Pexin paradust: well the "main" problem I had back then was they had broken the emulator on hardware without some cpu feature I lacked, and fixing it was not on their priority list (because who cares about old weak hardware anyway) 07:00 Pexin I don't remember details now 07:01 Pexin as I recall, software emulation refused to even attempt to work 09:48 MTDiscord https://cdn.discordapp.com/attachments/747163566800633906/974246657870954516/Screenshot_from_2022-05-12_02-45-30.png 09:48 MTDiscord victory 09:51 MTDiscord hooray 10:05 sfan5 was it the font rendering or did you have to fix something about the emulator? 10:11 Zughy[m] Wuzzy: hey man, can you please label your own issues and PRs, since you're a triager as well? It's the fourth time I have to do that for you, I don't think it's that fair.. 10:11 paradust sfan5: I just had to switch from ogles1 to ogles2 irrlicht driver :/ 10:12 Zughy[m] (I'm talking about things opened in these days) 10:12 paradust Is the ogles1 driver actually known to work? 10:12 sfan5 on phones yea 10:13 sfan5 since it's the default(?) 10:13 paradust before the last big merge? 10:13 sfan5 no idea 10:13 paradust i started reading through it and my mouth dropped 10:14 sfan5 Zughy[m]: I think they are unable to do this because the labeling interface requires javascript 10:14 paradust this is some terrible code: https://github.com/minetest/irrlicht/blob/master/source/Irrlicht/COGLESDriver.cpp#L585 10:15 paradust when was the last android release made? 10:15 sfan5 the last where things worked (roughly speaking) was 5.4.2, the actual last was 5.5.0 10:16 Zughy[m] sfan5: I hope that's not the reason, since 33% of the triager role is labelling issues (and I haven't seen him closing issues nor confirming bugs, so the left 66%) 10:17 sfan5 android before 5.5 used some random a bit patched irrlicht svn version (the main branch has no ogles support) 10:17 sfan5 and the rollout of 5.5 on google play was stopped because of sudden higher error rates (I think you're aware) 10:18 paradust do you have an exact repo/revision # for the irrlicht build used for the 5.4.2 and 5.5.0 android builds? 10:19 Zughy[m] sorry to interject, can #11568 be added to the next meeting discussing points? I know I have access to the organisation, but as far as I understand only core devs are allowed to express their opinions in the meeting thread 10:19 ShadowBot https://github.com/minetest/minetest/issues/11568 -- Consider a separate repo for requesting and discussing new features 10:22 sfan5 5.4.2: probably this https://github.com/MoNTE48/Irrlicht/tree/7fdb63d3cc8304ed3c86486e1aef712a0402f601 10:23 sfan5 5.5.0: https://github.com/minetest/irrlicht/releases/tag/1.9.0mt4 10:23 sfan5 if you plan on checking the differences between those versions to find possible issues I don't think that's gonna work out 10:23 paradust that was my plan ... 10:28 paradust the ogles1 driver has barely changed between those two and the current version 10:28 paradust so i guess the problem isn't the driver, but maybe the emulation for gles1 is not good 10:29 sfan5 that sounds more likely to me 19:26 MTDiscord sfan5: Looking at the code TileLayer.isTransparent() is equivalent to that switch in http://sprunge.us/I8gb4y?diff 19:51 sfan5 thanks, I already added the commit to a PR 19:55 sfan5 Zughy[m]: for me Bugfix and Feature are mutually exclusive with Maintenance 19:56 sfan5 I guess we should define labels better so there aren't these weird disagreements where nobody knows whats right 19:56 sfan5 also why is everything "Code quality" 19:57 sfan5 hyperbole of course but there's a few uses where I'm not sure what it's referring to 20:45 Zughy[m] sfan5: I was suggesting to ruben earlier to add an "@ Internal" component, for most of the issues present in the first page here: 20:45 Zughy[m] https://github.com/minetest/minetest/issues?q=is%3Aissue+is%3Aopen+-label%3A%22%40+Builtin%22%2C%22%40+Client+rendering+engine%22%2C%22%40+Devtest+Game%22%2C%22%40+Network%22%2C%22%40+Mainmenu%22%2C%22%40+Documentation%22%2C%22%40+Content%2FPkgMgr%22%2C%22%40+Translation%22%2C%22%40+Script+API%22%2C%22%40+Client+%2F+Controls+%2F+Input%22%2C%22%40+Build%22%2C%22%40+Client+%2F+Audiovisuals%22%2C%22%40+Mapgen%22%2C%22%40+Client+Script+API%22 20:45 Zughy[m] %2C%22%40+Startup+%2F+Config+%2F+Util%22%2C%22%40+Server+%2F+Client+%2F+Env.%22+ 20:46 Zughy[m] because I honestly don't know what to put 20:49 sfan5 internal to what? 20:50 Zughy[m] Like, it's all stuff related to C++ coding but that doesn't belong to any specific area (mapgen etc). See for instance the Android crash reports and the erlehmann issues. Maybe "@ Internal" is not the best name 21:07 erle “the erlehmann issues” are, for the most part, mishandling of some kind of input at the boundary between components. i would assign it to the component that consumes that input by default. 21:08 erle i fail to see though why anything *needs* a component 21:24 erle Zughy[m] asked me to replicate #5790 and oh wow, yes, it is extremely slow. anyone else has these wants to try it? https://mister-muffin.de/p/MNvk.txt 21:24 ShadowBot https://github.com/minetest/minetest/issues/5790 -- minetest.clear_craft unbearable slow when using the recipe as "identifier" 21:25 erle 2022-05-12 23:22:26: [Main]: time to clear 2048 crafts based on output: 0.01529 21:25 erle 2022-05-12 23:22:53: [Main]: time to clear 2048 crafts based on recipe: 24.10416 21:32 Zughy[m] erle: not having a component is a nice way to lose track of an issues for years 21:32 Zughy[m] *issue 21:32 erle well then the app should be neatly cut into components 21:33 erle Zughy[m] btw, what values do you get? are they similarly absurd as my 24 seconds to delete 2048 crafts? 21:33 erle i bet this is an extremely inefficient deep compare or so 21:40 Zughy[m] even worse, 28 seconds :D 21:41 Zughy[m] would you be so kind to post your script in the issue, reminding that is still a problem after 5 years? 21:41 Zughy[m] erle: 21:45 erle i'm not logged into github rn and they require me to wait for an email to log in nowadays (“security code”), which can take minutes or so, so if you are logged in, it would be faster to post it yourself (and mention me as the author ig) 21:46 erle ig i should up my 2FA game 21:46 erle if you don't do it, i'll try to remember and do it after sleep. good night! 22:26 MisterE[m] We should also add more ways to compare crafts to delete them