00:37 < bridge> LaTeX is quite enjoyable except for the very bad documentation for the core, I've been creating my CV in it and the ecosystem around it is quite flexible 00:37 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1208920263945887794/CV2.png?ex=65e509bb&is=65d294bb&hm=5fb98511fdc1cc74e594f251534de088900b1fe7fd1bfad75007106d54a1c071& 00:41 < bridge> Woah lerato cv 00:43 < bridge> I love latex but I never bothered to learn it so my cv and all other documents are the first template I find online lol 00:43 < bridge> 00:43 < bridge> For me it’s just a way to replace word with vim and track stuff in git. 00:45 < bridge> I'd only ever used latex in math mode until today, it's actually pretty enjoyable for text too 00:46 < bridge> Anything is more enjoyable than the other document creation tools out there 00:46 < bridge> Oh wait. Markdown is nicer :p 00:46 < bridge> Markdown to pdf is op 00:46 < bridge> Ngl WYSIWYG editors are just sooooo annoying. I made my first CV in libreoffice writer and I hated the experience so much 00:47 < bridge> Yes libreoffice is sadly not better than the proprietary stuff 00:47 < bridge> This programming like approach to designing a page is much more suitable to me 00:47 < bridge> Agree 00:47 < bridge> Wtf is WYSIWYG xd 00:48 < bridge> I can change the accentcolor in like one second and since I calculate all the other colors off of that it's instant 00:48 < bridge> Ah wait 00:48 < bridge> What you see is what you get 00:48 < bridge> What You See Is What You Get 00:48 < bridge> Dude I’m good 00:48 < bridge> Holy 00:48 < bridge> You are too smart for this world 00:49 < bridge> Go to bed already lerato 00:49 < bridge> It’s Monday 00:50 < bridge> I'm truly a NEET nowadays, actualy literally nothing to do tomorrow 00:50 < bridge> Okay I’m having s hard time cracking that one 00:51 < bridge> Well that's a lie, I did promise I'd tutor someone tomorrow, but it's in the afternoon, so no need to wake up early, and he doesn't pay me so it doesn't constitute employment 00:51 < bridge> Which country are you in this week? 00:51 < bridge> Italy for a bit now, though I'm very heavily considering going to japan next week 00:52 < bridge> I found an insane flight deal 00:52 < bridge> Classic lerato 00:52 < bridge> So it’s like 1 in the morning for you? 00:52 < bridge> yep 00:52 < bridge> Okay thats still ok 00:52 < bridge> So what’s a NEET 00:52 < bridge> I'll play a game of league then sleep. I earned one game today, I worked hard on my CV, studied japanese, added one more component to my vulkan experiment 00:53 < bridge> Eww lol 00:53 < bridge> The one with the Chinese state hacker kernel level root kit? 00:53 < bridge> NEET is "Not in Education, Employment, or Training". The people wasting our societies resources 00:53 < bridge> Not yet 😄 00:54 < bridge> Will you continue playing? 00:54 < bridge> Does it even work on Linux? 00:54 < bridge> Or do you use Mac idk 00:54 < bridge> Yes I will. Yes it does work on linux (but it won't when the rootkit gets added) 00:54 < bridge> I dualboot into windows just for a handful of games including league 00:55 < bridge> Oh no we lost lerato 00:55 < bridge> I pretty much don't care, everything on this machine is encrypted and there is nothing interesting on the windows partitions that aren't 00:55 < bridge> Booting into windows just to run a rootkit sounds so sad 00:56 < bridge> When rewrite lol in ddnet engine 00:56 < bridge> Xi can watch me edit my holiday photos in lightroom or watch me play league 00:56 < bridge> Send holiday photos 00:56 < bridge> I guess he can also see my discord, which is sadge but whatever. Nothing really interesting in here either except me slowly going insane 00:57 < bridge> Health insurance bill just increased by 10% because of that msg :p 00:57 < bridge> He would never do that. I support the CCP wholeheartedly 00:57 < bridge> lmao 00:57 < bridge> Me too me too 00:57 < bridge> discord users be like 00:58 < bridge> Here is a pic I took in glasgow 00:58 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1208925469957693470/DSC02076.jpg?ex=65e50e94&is=65d29994&hm=09a69bafc4bb5a9202efe3f15c9c774bfb59650a857991d07f98b762420d447c& 00:58 < bridge> AI generated! 00:58 < bridge> I actually don't remember if this is one where I removed a dude or the other 00:59 < bridge> Why xd 00:59 < bridge> I can see the normal map on the asphalt! 00:59 < bridge> Why bother to edit 01:00 < bridge> I'm going to sleep, maybe some anime first. Some nice new episodes came out this weekend. Good night my brain is fried 01:01 < bridge> Well wdym by edit? Adjusting colors and exposure or the heavier stuff like removing people? 01:01 < bridge> Both 01:01 < bridge> I wouldn’t do any of it 01:01 < bridge> Especially if I’m not posting them 01:01 < bridge> Well they look better edited. Editing is part of the art of photography 01:01 < bridge> I'm posting them on instagram 😄 01:02 < bridge> Or do you have some travel blogger social 01:02 < bridge> Oh 01:02 < bridge> Send 01:02 < bridge> Influencer lerato 01:02 < bridge> Find it yourself, all the required info is out there 😄 01:05 < bridge> @Lerato2 01:06 < bridge> That is tempting but I’m not booting instagram for that 01:06 < bridge> Anyways good talk I’ll go make some coffee now c: 05:47 < bridge> masterserver not validate json info from servers? 05:47 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1208998247675199561/image.png?ex=65e5525c&is=65d2dd5c&hm=ea143de87dd9be032f0d48f128891993733f2e400e5761bfb3ea338d51326dc6& 05:47 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1208998247935516713/image.png?ex=65e5525c&is=65d2dd5c&hm=e75dff8e8ec1b712bb66d2b30c51d298dc0b987fe65b493999e57ec2511002f2& 05:48 < bridge> :thonk: 05:53 < bridge> Hello guys, I'm creating a documentation that aims to be linear for the teeworlds protocol. You can access via https://protocol-docs.teeframe.com/. Its very incomplete and any kind of help is welcome (mainly for connless packets as I know nothing about them xd) 06:08 < bridge> lol 06:08 < bridge> 2024-01-30\13_08_50.json 06:08 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1209003475057377320/image.png?ex=65e5573a&is=65d2e23a&hm=b564899c3de549b1987a18966f059714a170752f926bfeb400475d8714b2d8a9& 06:15 < bridge> I think chillerdragon alr have made this 07:24 < ws-client> @reitw ye thats where he copy pasted it from lmao 10:00 < bridge> Nah you can send whatever you want 12:00 < bridge> cringe 12:28 < bridge> https://chillerdragon.github.io/teeworlds-protocol/index.html 12:50 < bridge> trol? :lol: 12:52 < ws-client> i love how this repo has more stars than the total amount of stars i made by fulltime githubbing for like 5 years https://github.com/klonnet23/helloy-word 13:38 < bridge> i need placement new in rust 13:38 < bridge> 13:38 < bridge> i ran into stack overflow during a intialization :lol: 13:38 < bridge> now i create way to many smaller boxes for nothing 13:38 < bridge> today is "i hate programming"-monday 13:39 < bridge> yea 13:57 < bridge> You can construct a box from a raw pointer in an unsafe block, is that not enough? 14:02 < bridge> Yeah but don't want to introduce more unsafe. I guess I can live with a little performance overhead for now 14:03 < bridge> Understandable 14:03 < bridge> How about nightly `new_in` instead? 14:08 < bridge> Yet I haven't introduced any nightly features, but sometimes I really consider it xd 14:08 < bridge> I really hate rust analyzer. It's so slow and so buggy.. Defs the worst in all rust ecosystem 14:22 < bridge> Huh, I've had a much better time with it then I've had with clangd. But I only dabble in the most vanilla of rust features 14:25 < ws-client> @miguilim this one might come in handy for you 14:25 < ws-client> https://private-user-images.githubusercontent.com/20344300/305960917-2fe66dc9-564a-4e54-b94d-a0cdc56b187e.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3MDgzNDkzNzEsIm5iZiI6MTcwODM0OTA3MSwicGF0aCI6Ii8yMDM0NDMwMC8zMDU5NjA5MTctMmZlNjZkYzktNTY0YS00ZTU0LWI5NGQtYTBjZGM1NmIxODdlLnBuZz9YLUFtei1BbGdvcml0aG09QVdTNC1ITUFDLVNIQTI1NiZYLUFtei1DcmVkZW50aWFsPUFLS 14:25 < ws-client> UFWQ09EWUxTQTUzUFFLNFpBJTJGMjAyNDAyMTklMkZ1cy1lYXN0LTElMkZzMyUyRmF3czRfcmVxdWVzdCZYLUFtei1EYXRlPTIwMjQwMjE5VDEzMjQzMVomWC1BbXotRXhwaXJlcz0zMDAmWC1BbXotU2lnbmF0dXJlPTM0NzllY2RhNjk4YTlhNWQyMjVhODM4OTgxYjE2ZjVhNzk5YjA0NzZkYzRlNGIyMGFiM2Q5ODU5YTA1ZGZhNWEmWC1BbXotU2lnbmVkSGVhZGVycz1ob3N0JmFjdG9yX2lkPTAma2V5X2lkPTAmcmVwb19pZD0wIn0.lxvKir5X0pvvwmSRwboBKPlrrM8ett46vs8Kbu_nutE 14:26 < ws-client> https://twnet.zillyhuhn.com/?d=04+0a+00+cf+2e+de+1d+04 14:31 < bridge> sploink 14:44 < bridge> Yeah dunno, I've rather bad experience way too often. 14:44 < bridge> 14:44 < bridge> - It's not pretty well tested. The nighty version often is broken. 14:44 < bridge> - using git pull often requires a cargo clean because it still sees outdated function signatures 14:44 < bridge> - it's often much slower than a cargo build. I struggles hard getting dependencies together. 14:44 < bridge> - I have it on analyze on clicking save, but very often it takes way to long to get insight of the code, e.g. type information of variables. This one is rather random, sometimes it works well, sometimes not 14:46 < bridge> Rustfmt is the second worst BTW xd, it fails at so random stuff. Especially annoying is inside macros 14:48 < bridge> i mean its nightly for a reason xd 14:48 < bridge> but ye rust analyzer is hard 14:48 < bridge> the previous was rly rly rly worse 14:48 < bridge> rustls 14:48 < bridge> That's true, but for something so essential for a ecosystem I'd expect them to have much better testing 14:48 < bridge> i havent noticed any big issues myself tho 14:49 < bridge> im sure nightly means nightly like ddnet nightly 14:49 < bridge> idk its probs hard to test this 14:49 < bridge> so many combinations 14:49 < bridge> but dont worry edlang language server will have ur back 14:49 < bridge> Yeah I assume so too, but aren't they paid devs? They can take time to think about testing methods 14:49 < bridge> in 2040 14:50 < bridge> Xd 14:50 < bridge> idk if they paid 15:03 < bridge> ChillerDragon btw edlang has 4 stargazers xD 15:03 < bridge> https://github.com/edg-l/edlang 15:03 < bridge> its coming along nicely 15:03 < bridge> need to add arrays 15:09 < bridge> @ryozuki heeeeeeeeelp 15:09 < bridge> ``` 15:09 < bridge> Missing request extension: Extension of type `alloc::sync::Arc>` was not found. Perhaps you forgot to add it? See `axum::Extension`. 15:09 < bridge> ``` 15:09 < bridge> 15:09 < bridge> the pool is accessible to all route handler's using `.with_state(pool.clone())` 15:09 < bridge> meaning `Arc` should be available to all handlers via `State` right?? RIGHT??? 15:14 < bridge> `State(pool): State>,` instead of `Extension(pool): Extension>,` 15:16 < bridge> itsb een long since i used axum but 15:16 < bridge> ye use state not extension 15:16 < bridge> i told u before 15:16 < bridge> you did? 15:16 < bridge> or to someone xd 15:16 < bridge> state is type checked at compile time 15:16 < bridge> so its better 15:21 < bridge> <3 16:57 < bridge> @chairn https://www.youtube.com/watch?v=1CSeY10zbqo 17:48 < bridge> @jupeyy_keks https://wccftech.com/amd-zen-5-cpus-reportedly-made-on-tsmc-3nm-process-q2-mass-production-q3-2024/amp/ 17:48 < bridge> 3nm???? 17:48 < bridge> > AMD Zen 5 CPUs Reportedly Made On TSMC’s 3nm Process In Q2, Mass Production In Q3 18:06 < bridge> I thought they will use zen 4 for now 18:06 < bridge> That would be nice news 21:23 < bridge> ChillerDragon: perhaps you can find a better matrix client than element 🙂 21:28 < bridge> read-only only applies to the contents on linux 21:31 < bridge> I'm not sure if that has been tested in years. probably simply not working. is there a particular reason why you want to use mysql with ddnet on windows? 21:31 < bridge> it was something i was dependent on on a modification i am working on 21:32 < bridge> but since i just dumped the entire thing and wrote a restapi - i dont need it anymore 21:32 < bridge> kekw 21:32 < bridge> it was something i was dependent on for a modification i am working on 21:44 < bridge> I also think that lua might be the better pick right now, but I also don't have a lot of experience working with wasm 21:45 < bridge> that sounds like a bad idea. the user doesn't really have the context to make this decision IMO 21:49 < bridge> good the opposite of what you say is usually the way to go 21:50 < bridge> the only reason u think that is bcs factario is doing it 21:51 < bridge> LUA files sent by a server can crash or "hack" (steal the personal data) the clients, and I don't see a real workaround here. We can't analyze LUA for bugs. Maybe we could install some handlers and interceptions on the client side but it still looks unreliable and tricky. 21:51 < bridge> (So for now I want to use LUA for the server side scripting but don't have a solution for client-side mods) 21:52 < bridge> "bcs factorio" and tens other games "is doing it" 21:52 < bridge> and maybe bcs there is approximately zero games using wasm for that. 21:52 < bridge> yes it's very hard to do that.. even browsers fail sometimes 21:53 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1209241368157888542/image.png?ex=65e634c8&is=65d3bfc8&hm=12fce2a29e7f4e8b2d699591c02680845b6451cfdba710e4592ceddb5afafc26& 21:53 < bridge> that doesn't make lua better than wasm tho. 21:53 < bridge> 21:53 < bridge> browsers will use more and more wasm, so it's generally more likely to contain less security related bugs. 21:53 < bridge> Also there are game engines that start using wasm 21:53 < bridge> also wasm, as i imagine it, will allow lua 21:54 < bridge> bcs it's a superset 21:54 < bridge> Bcs it is a different tool for a different purpose. 21:55 < bridge> anyway, more secure + faster = the future for me 21:55 < bridge> i'm biased 21:55 < bridge> Would you say that a C++ compiler a superset of a web browser? 21:55 < bridge> You can do a web browser using C++ compiler but (normally) not vice versa. 21:55 < bridge> Would you say that a C++ compiler is a superset of a web browser? 21:55 < bridge> You can do a web browser using C++ compiler but (normally) not vice versa. 21:56 < bridge> i don't really understand it, but i guess i'll just answer with yes 21:56 < bridge> well maybe they are equally poweful, since a browser can run a c++ compiler compiled to wasm :lol: 22:00 < bridge> @zenopeer @learath2 you need to support the extended server info (the non `FSTD`/`DTSF` one) for the new masterserver's backcompat. alternatively register via HTTPS, it should be easier™ 22:04 < bridge> amazing 😮 22:10 < bridge> Yep, you're answering yes. 22:10 < bridge> Compiler and Web browser are different tools for different purposes. I need a web browser and you're selling me a "secure + faster" compiler, because "after all I also get a web browser" using it. 22:12 < bridge> to me wasm is exactly what LUA can do + even more 22:12 < bridge> why is it a different tool for you? 22:12 < bridge> the way i use wasm is not the way you __have__ to use it 22:12 < bridge> in the browser wasm is used interchangeable with js 22:13 < bridge> ChillerDragon: default argumets are evaluated once when the source file is read IIRC. it's not "whatever the first instance of MsgClSay had" 22:16 < bridge> can you show the `PACKET_GETINFO` constant? 22:18 < bridge> bro when u work on smth for 4h and it works first try 22:18 < bridge> 22:18 < bridge> epyc 22:18 < bridge> 22:18 < bridge> i have the feeling that is easier with rust 22:18 < bridge> 22:18 < bridge> @ryozuki i already have to do the propaganda lately, when do you take your job back 😬 22:20 < bridge> because `map` guarantees that the function is executed even for `skip`ped elements 22:20 < bridge> heard there's a whole book on tex 22:21 < bridge> mh, but i think llvm optmized it still. i hoped to get perf uplift in release mode 22:21 < bridge> do tell 22:22 < bridge> correct, the info is a free form JSON object 22:23 < bridge> `new_in` doesn't look like placement new 22:23 < bridge> I still think intellij's rust support is quite nice 22:23 < bridge> yea, that's really annoying. I considered starting a fork for that shit 😦 22:24 < bridge> 😦 22:25 < bridge> in that sense, lua also allows wasm 22:25 < bridge> i mean right now 22:25 < bridge> not theoretically 22:26 < bridge> I bet there's a wasm interpreter in lua 22:26 < bridge> right noww 22:26 < bridge> why is there no strong typing? 22:26 < bridge> I'll look it up though 22:26 < bridge> what? never xd 22:26 < bridge> so people can add their own stuff 22:26 < bridge> how slow should that be? 22:27 < bridge> probably quite fast if you use luajit 22:27 < bridge> is luajit memory safe? 22:27 < bridge> or better: will the wasm runtime be memory safe 22:30 < bridge> well, let them add it to separate fields, but the fields common to all servers must be validated 22:30 < bridge> yes, just like lua 22:30 < bridge> is it written in c? 22:30 < bridge> like lua 22:30 < bridge> I basically modeled it after the old server info. servers could also send all kinds of data in the server info 22:30 < bridge> in some archives stats im found server without "name" field 😄 22:30 < bridge> yes, although for JITs, the main unsafety doesn't really come from being written in C, I think 22:30 < bridge> sure. the client will just disregard that 22:30 < bridge> but rather from the fact that wrong code generation itself can lead to security vulnerabilities 22:31 < bridge> i cant find the lua benchmarks, but luajit also wasnt that fast 22:31 < bridge> i need ryo's 2000 tabs 22:34 < bridge> wdym 22:34 < bridge> okay, I'll have to write my own parser for this, instead of one line solution: 22:34 < bridge> `JsonSerializer.DeserializeAsync(entry.DataStream!)` :justatest: 22:34 < bridge> I think it outperformed V8 22:34 < bridge> is that a skill? xd 22:34 < bridge> they also had the case where they were faster than C for calling into dynamic libraries 😄 22:34 < bridge> 430 euros for a return flight 22:34 < bridge> yes, just like you'd need to if you received the server infos from the servers directly 22:35 < bridge> Can't you do it with a convoluted allocator 22:35 < bridge> it gets the new thing as a by-value parameter 22:35 < bridge> you've already lost at that point 22:35 < bridge> Ah, a shame 22:35 < bridge> https://programming-language-benchmarks.vercel.app/lua-vs-wasm 22:36 < bridge> rust code compiled to wasm 22:36 < bridge> vs luajit 22:36 < bridge> hello world rust looses.. as expected :lol: 22:36 < bridge> also note wasmtime isnt the fastest runtime 22:38 < bridge> luajit seems to be 10%-30% slower than wasmtime based on these tables, if I read them correctly 22:38 < bridge> + u need to learn lua 22:39 < bridge> otherwise you'll probably need to learn some wasm stuff if you want to use a language that's not yet explicitly added? 22:40 < bridge> yes, but learning rust is not a waste of time 22:40 < bridge> why are u even fighting so hard for lua xD 22:40 < bridge> but for the advantage of compiling lua to wasm, you'd first have to figure out how to pass parameters, export functions, etc. 22:40 < bridge> u never used it yourself, u only saw it in factario 22:40 < bridge> is that so convincing? 22:41 < bridge> and why would you not need to do that for c++ or rust? 22:41 < bridge> I saw it embedded in tw, probably embedded it into my own project at some point, worked a bit with it 22:41 < bridge> wasm = your rust or c++ code 22:42 < bridge> and I'm only trying to make the response on this server less biased 22:42 < bridge> for which sense? 22:42 < bridge> if you'd be preaching that lua was the only possible solution, I'd probably talk about wasm a bit 22:42 < bridge> backward compability is so FUCKING AWESOME 22:42 < bridge> because I don't think that wasm is the only possible solution here, or that it's necessarily the best 22:42 < bridge> i love it 22:42 < bridge> man, it's so amazing using protol ex 22:43 < bridge> please stop mocking me… 22:43 < bridge> #7993 22:43 < bridge> https://github.com/ddnet/ddnet/pull/7993 22:43 < bridge> i try to see if u take the opposite side 22:43 < bridge> ah, I misunderstood it 22:44 < bridge> if you present arguments for backward compatibility, I can tell you counter arguments 🙂 22:44 < bridge> you didn't say "wasm is so awesome" 22:44 < bridge> you provided arguments 22:44 < bridge> backward compability makes you read code harder which ultimately makes you know the code base better 22:45 < bridge> that sounds not very convincing. why would it be useful if you had to concentrate more to read code? 22:45 < bridge> the hardest challenges make the best ppl 22:45 < bridge> I disagree 22:45 < bridge> it's better if you can quickly jump in and start coding 22:46 < bridge> more contributors, more fresh blood, etc. 23:58 < bridge> Yes? V8 is an insanely commonly used js engine written by google, with jit, outperforming it is actually impressive