00:35 < bridge> https://www.youtube.com/watch?v=Mc_JCFUzbXA 00:36 < bridge> i was mad when he was presenting but actually i think he's right 01:10 < bridge> hello 01:10 < bridge> https://github.com/Siile/KillingFloor How do I compile this on Linux? 01:10 < bridge> I forgor💀 01:10 < bridge> I wanna test sum mods ppl uploaded on gh 01:41 < bridge> compiling_everything_linux 01:41 < bridge> https://www.teeworlds.com/?page=docs&wiki=compiling_everything_linux 01:55 < bridge> alr tyyy 06:56 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200318258306351164/image.png?ex=65c5be7d&is=65b3497d&hm=ed10edc14643a9903c1971a5126dc7409e50719ed26d1b48d1b5c0a89aab7045& 06:56 < bridge> it still says 2022 06:56 < bridge> lmfao 06:56 < bridge> who wanna take a funny picture with me 09:42 < bridge> gm 09:43 < bridge> morning 10:05 < bridge> idk where to put myself in the thing... 10:48 < bridge> I'm down right hanging off of a tee on the bottom of the island 🤫 11:19 < ws-client> Monthly shoutout to @Patiga for building twmap <3 11:19 < ws-client> https://zillyhuhn.com/cs/.1706264321.png 11:20 < ws-client> Being able to quickly parse digits placed by the mapper and use them in a script is so cool! 13:09 < bridge> Did you see `std::linalg` is coming in C++26? 😄 13:12 < bridge> cool 13:47 < bridge> chillerdragon: r u twmap pro? 13:47 < bridge> i have a request: 13:47 < bridge> i want to know the map with the most layers in a single group 13:48 < bridge> ok no, i actually want the most tile count of the whole group.. so the sum of all tile layers in a single group 13:48 < bridge> ah and only the theoretical tile count, so tiles with index 0 count too 13:49 < bridge> max of (group w * group h * tile layer count in group) 14:28 < bridge> welcome to friday rust 14:28 < bridge> the crates of the day 14:28 < bridge> - https://docs.rs/internment/0.7.4/internment/index.html 14:29 < bridge> https://docs.rs/hashconsing/1.6.0/hashconsing/ 14:33 < bridge> https://inference-review.com/article/the-man-who-carried-computer-science-on-his-shoulders 14:40 < bridge> I've seen this before and wonder if it's ever useful for anything other than a static str 14:41 < bridge> i recommend reading this, its extremely interesting 14:41 < bridge> its specially useful for compilers 14:42 < bridge> @jupeyy_keks rust MIR uses interning everywhere 14:42 < bridge> for basic blocks, and all mir structures 14:42 < bridge> like types 14:42 < bridge> etc 14:42 < bridge> K 14:43 < bridge> I wanted to use it for efficient indices. But isn't better than simply using a hashmap. In fact even worse xd 14:43 < bridge> Chatgpt baited me 14:45 < bridge> @jupeyy_keks this script is rly close to what you want 14:52 < bridge> Nice 15:04 < bridge> I am fully booked atm sorry 15:05 < bridge> it's fine 15:05 < ChillerDragon> did patiga attach a script? i see neither a reply on matrix nor irc atrifacts 15:05 < bridge> i have to touch pyson 15:05 < bridge> my nightmare 15:05 < ChillerDragon> xd 15:05 < bridge> he replied to a previous post where he attached a script 15:06 < ChillerDragon> too old for matrix maybe 15:06 < bridge> https://discord.com/channels/252358080522747904/293493549758939136/999263776140447794 15:06 < bridge> if that helps 15:42 < bridge> @jupeyy_keks when publish hiarc 15:42 < bridge> make ur lib correct: https://rust-lang.github.io/api-guidelines/checklist.html 15:42 < bridge> you have a great taste 😬 15:43 < bridge> WTF 15:43 < bridge> who reads all these 15:43 < bridge> me 15:43 < bridge> most is common sense 15:43 < bridge> but an important one is to eagerly implement derives 15:43 < bridge> all u can 15:43 < bridge> in libs 15:44 < bridge> Types eagerly implement common traits (C-COMMON-TRAITS) 15:44 < bridge> 15:44 < bridge> Copy, Clone, Eq, PartialEq, Ord, PartialOrd, Hash, Debug, Display, Default 15:44 < bridge> yeah i agree with all except Default 15:44 < bridge> u dont know how annoying it is to use a lib with structs that dont implement for example Debug when they cn for free 15:45 < bridge> @jupeyy_keks eagerly for me means implement them if a simple derive works and it doesnt make it error prone 15:45 < bridge> if default rly makes no sense then dont 15:45 < bridge> i see 15:45 < bridge> but except 2 structs hiarc is really only a trait + derive macro 15:45 < bridge> > Conversions use the standard traits From, AsRef, AsMut 15:45 < bridge> but what sucks is implementing the trait for all std types 15:46 < bridge> this is probably the biggest thing that holds me back xd 15:46 < bridge> https://zed.dev/ anyone tried this yet? it's open source and made in rust 15:46 < bridge> no 15:46 < bridge> bro we mentioned this yesterday 15:46 < bridge> does it have debugger support? 15:46 < bridge> its macos only 15:46 < bridge> i tried it 15:47 < bridge> it looks nice 15:47 < bridge> macos only... 15:47 < bridge> I'll definitely hate it if the developers are mac enjoyers, will pass 15:47 < bridge> :lol: 15:47 < bridge> but true 15:48 < bridge> who the fuck develops using metal 15:48 < bridge> instead of vulkan 15:48 < bridge> mac enjoyers 15:49 < bridge> ye its stupid 15:49 < bridge> the editor looks neat and responsive 15:49 < bridge> its written in rust 15:49 < bridge> its idiocy its only macos 15:49 < bridge> i want to know if it has debugger support 15:49 < bridge> else it's just a text editor, not better than lapce probably xdd 15:50 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200452584662646905/image.png?ex=65c63b97&is=65b3c697&hm=72b0afa8269b7f21a4fef79ecfbaf89b05b609168ecaa093871c8ba8411cbc85& 15:50 < bridge> How is lapce? 15:51 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200452983431905331/image.png?ex=65c63bf6&is=65b3c6f6&hm=213caea8ee9879f3b01c70c428cdc29f627b0b7fb6de2d34e0171cded8bd0069& 15:51 < bridge> pros: 15:51 < bridge> - very fast (esp the rust analyzer integration) 15:51 < bridge> cons: 15:51 < bridge> - no debugger support 15:51 < bridge> - it still feels worse as a text editor than most text editors, moving selected text doesnt work etc. 15:51 < bridge> I've been thinking of migrating to something new, I'm kinda feeling neovim but exploring other stuff 15:52 < bridge> but i used it 3 months ago 15:52 < bridge> i should try today 15:52 < bridge> neovim rocks 15:52 < bridge> but i am 99% sure they still dont have debugger integration 15:52 < bridge> and that is simply a red flag for me 15:52 < bridge> no debugger support is completely fine for me, I've never had debugger support anyway 15:52 < bridge> me neither xd 15:52 < bridge> gdb 15:52 < bridge> nani 15:52 < bridge> gdb per hand is way too slow to use xD 15:52 < bridge> try gdb with gef 15:53 < bridge> I've gotten so used to it that I honestly don' 15:53 < bridge> 't even feel slow anymore 15:53 < bridge> we can host a competition 15:53 < bridge> no we can't, you are far more in shape than me 15:53 < bridge> I haven't written code in months 15:54 < bridge> If I truly need to see my code while debugging exactly at the line I'm at, I'll use the tui 15:55 < bridge> but honestly don't even remember last I activated the tui, I just have the code up on the window next to me 15:55 < bridge> well it's also about watchers and thread lists etc for me 15:55 < bridge> local variables 15:55 < bridge> hovering over variables 15:56 < bridge> i dunno, i mean i also often do log driven debugging 15:56 < bridge> watchers and thread list I can have with the cli quickly 15:56 < bridge> but sometimes a debugger is simply nicer.. and a good GUI enhances that 15:56 < bridge> hovering over local variables would be great though ngl 15:56 < bridge> I mean I don't mind doing a quick `p` but still it'd be better UX 16:00 < bridge> lapce 16:00 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200455107444219964/image.png?ex=65c63df0&is=65b3c8f0&hm=7abe0963938c24f3ab380dc3cbf1f54778efbcc4a96d299f24cfa82982e4f57b& 16:00 < bridge> feels already closer to vscode since i used it last time 16:00 < bridge> but still feels worse overall xd 16:01 < bridge> what's it do? 16:01 < bridge> Linear Algebra 16:02 < bridge> So we can have matrix multiplication without having to get a library 16:02 < bridge> (or doing it by hand :D) 16:04 < bridge> I wonder how you design that in a general purpose fashion 16:04 < bridge> does it use indices? 16:04 < bridge> https://en.cppreference.com/w/cpp/header/linalg 16:04 < bridge> execution policy 16:05 < bridge> template __inout_vector InOutVec1, __inout_vector InOutVec2, class Real> 16:05 < bridge> void apply_givens_rotation(ExecutionPolicy&& exec, 16:05 < bridge> InOutVec1 x, InOutVec2 y, Real c, Real s); 16:05 < bridge> xd 16:05 < bridge> for cuda? 16:05 < bridge> ```cpp 16:05 < bridge> template __inout_vector InOutVec1, __inout_vector InOutVec2, class Real> 16:05 < bridge> void apply_givens_rotation(ExecutionPolicy&& exec, 16:05 < bridge> InOutVec1 x, InOutVec2 y, Real c, Real s); 16:05 < bridge> ``` 16:05 < bridge> looks horrible to use, ngl xD 16:10 < bridge> There is a signature that doesn't require an execution policy too 16:10 < bridge> ```c++ 16:10 < bridge> template<__in_matrix InMat1, __in_matrix InMat2, __out_matrix OutMat> 16:10 < bridge> void matrix_product(InMat1 A, InMat2 B, OutMat C); 16:10 < bridge> ``` 16:10 < bridge> Idk, it doesn't look particularly awful to use, just ignore the templating 16:10 < bridge> how does it access the numbers? 16:12 < bridge> No idea, I don't even know if there is any implementation of it tbf 16:12 < bridge> Let me take a look at the proposal 16:17 < bridge> It seems to be through accessors functions, why? 16:21 < bridge> Oh, nope, it's accessed directly with offsets it seems, you need to use a "BLAS-compatible memory layout" for your vectors and matrices 16:22 < bridge> does that mean it's slices with optional stride? 16:24 < bridge> I think so, yes. It's a little hard to read through a proposal this quickly but that's what it seems to be 16:25 < bridge> A matrix/vector is basically an `std::mdspan`, which is like a slice with optionally a stride 16:51 < bridge> afair vsc has a vim motion plugin, nice getting used to it while still having the possibility to dodge to mouse if needed - but after getting used to neovim once any other IDE just feels soooo slow :D 16:51 < bridge> That's what I'm using, currently, but I feel it's time to do something new 16:51 < bridge> learn emacs :justatest: 16:51 < bridge> jkjk 16:52 < bridge> I actually never tried emacs, hm 16:52 < bridge> :D oh no 16:52 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200468420932681819/image.png?ex=65c64a57&is=65b3d557&hm=656e2abf0477da8501d6ff1a40cb6b20aa014d215b2e7c7782fb3eff45cdd663& 16:53 < bridge> u should join a emacs discord. i bet they convert you to a emacian 16:53 < bridge> have u tried emacs OS? 16:53 < bridge> xd 16:53 < bridge> emacs as PID1 16:53 < bridge> hvae u tried emacs on your washing machine 16:53 < bridge> why not try an nvim pre config? there are some nice one's out there: "nvchad, lunarvim, lazyvim" 16:54 < bridge> I don't do that 16:54 < bridge> they have important things preconfigured (treesitter, LSP, filetree) 16:54 < bridge> If I'm going to use neovim I'll do it myself 😄 16:54 < bridge> even tho :Ex exists ofc 16:55 < bridge> configuring vim from scratch was a lot of fun for me personally, i used it as a way to learn lua on the way 16:55 < bridge> and then i completly dumped my config and rewrote it in nix 16:55 < bridge> Or maybe I should go all the way back to where I started to vim 16:56 < bridge> go back to the roots of all things, use Vi 16:57 < bridge> vi is just too janky 16:57 < bridge> use ed 16:57 < bridge> https://en.wikipedia.org/wiki/Ed_(text_editor) 16:59 < bridge> > (In)famous for its terseness, ed, compatible with teletype terminals like Teletype Model 33, gives almost no visual feedback,[8] and has been called (by Peter H. Salus) "the most user-hostile editor ever created", even when compared to the contemporary (and notoriously complex) TECO.[2] For example, the message that ed will produce in case of error, and when it wants to make sure the user wishes to quit without saving, is "?". It does not report 17:07 < bridge> who even needs those fancy things called editors, just use `echo "blah blah" > file` :justatest: 17:08 < bridge> ```sh 17:08 < bridge> 17:08 < bridge> #!/bin/bash 17:08 < bridge> 17:08 < bridge> # Simple Bash Text Editor 17:08 < bridge> 17:08 < bridge> # Check if a filename was provided 17:08 < bridge> if [ $# -eq 0 ]; then 17:08 < bridge> echo "Please provide a filename." 17:08 < bridge> exit 1 17:08 < bridge> fi 17:08 < bridge> 17:08 < bridge> FILENAME=$1 17:08 < bridge> 17:09 < bridge> # Function to display the file content 17:09 < bridge> function view_file() { 17:09 < bridge> clear 17:09 < bridge> echo "Editing: $FILENAME" 17:09 < bridge> echo "--------------------------" 17:09 < bridge> cat $FILENAME 17:09 < bridge> echo 17:09 < bridge> echo "--------------------------" 17:09 < bridge> echo "Type 'SAVE' to save." 17:09 < bridge> echo "Type 'EXIT' to exit without saving." 17:09 < bridge> echo 17:09 < bridge> } 17:09 < bridge> 17:09 < bridge> # Main loop 17:09 < bridge> while true; do 17:09 < bridge> view_file 17:09 < bridge> echo "Enter your text (append mode):" 17:09 < bridge> leak of chillerdragons editor 17:09 < bridge> :poggers2: 17:09 < bridge> he probably uses irc as editor 17:21 < bridge> I have seen this guy use cat unironically as a editor casually in a coding video https://youtube.com/@awesomekling?si=ELlMlmlUO3Y6gtMV 17:28 < bridge> lmao the C++ integration into the Linux kernel discussion is back 17:28 < bridge> https://lore.kernel.org/lkml/3465e0c6-f5b2-4c42-95eb-29361481f805@zytor.com/ 17:52 < bridge> its pointless with rust 17:59 < bridge> well c is an 99.99% subset of cpp xd 18:00 < bridge> so it could be used for existing c code 19:29 < bridge> no video description, what is it about ? 19:34 < bridge> damn, some guys are really upset about this 😄 19:34 < bridge> 19:35 < bridge> damn, some guys are really passionate about this 😄 19:35 < bridge> 19:35 < bridge> damn, some guys are really passionnate about this 😄 19:35 < bridge> 20:18 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200520271417852054/2024-01-26_20-16-52.mp4?ex=65c67aa1&is=65b405a1&hm=061fbe39ee4c173ddd94b837c4f3923c51921f41f5bad0a7eb955997743762ce& 20:19 < bridge> the game doesnt really like 2 of itself being opened lol 20:19 < bridge> should be no problem 20:19 < bridge> only limit is your ram 20:20 < bridge> that cqwrteur should just be banned 20:20 < bridge> 32 gb not enough to run 2 ddnets? 20:21 < bridge> 32 gb should be plenty to run two copies of ddnet 20:22 < bridge> ok it doesnt happen on opengl 20:25 < bridge> sounds like it might be a bug with vulkan 20:26 < bridge> @cyberfighter when will u finally reinstall your os 20:26 < bridge> which 20:26 < bridge> u always have problems even after changing your setup 20:26 < bridge> i reinstalled it already 20:27 < bridge> im just best bug reporter 20:27 < bridge> :troll: 20:35 < bridge> ok i restarted my game after setting it to vulkan 20:35 < bridge> and the screen is completely black 20:36 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200524611125780521/2024-01-26_20-35-15.mp4?ex=65c67eab&is=65b409ab&hm=06bb95230b0120b3419423d954a8f116348a43d73cad037eb5aaf7e980daf233& 20:36 < bridge> just imagine that its ddnet client 20:37 < bridge> wdym I usually open multiple clients like 2+ is a standard on 16GB. 20:40 < bridge> damn fsr 3 is out since a week & it's horrible on cod lmao 20:41 < bridge> Should be fixed by #7859 and #7860 20:41 < bridge> https://github.com/ddnet/ddnet/pull/7859 20:41 < bridge> https://github.com/ddnet/ddnet/pull/7860 20:50 < bridge> ok 21:21 < bridge> seems to be working fine xd 21:21 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200535975080775850/2024-01-26_21-20-12.mp4?ex=65c68941&is=65b41441&hm=933a2e1a9a658635605162df4ec481f91041f7e2c8e09f24fedac3271ed67fdd& 21:26 < bridge> xd 21:30 < bridge> @robyt3 so it's blackscreen, what happened yesterday 21:31 < bridge> validation errors should close the game or not? 21:31 < bridge> yeah, I also still get the same validation errors with #7860 21:31 < bridge> https://github.com/ddnet/ddnet/pull/7860 21:31 < bridge> mh ok 21:32 < bridge> well guess i have to start windows tomorrow xd 21:32 < bridge> doesn't seem like it's fatal. 21:32 < bridge> `vulkan_debug: validation error: Validation Error` 21:32 < bridge> mhh ok 21:33 < bridge> I think #7859 is fine unrelated to this though, the client should definitely start in the foreground when using the restart button, which is fixed by that PR 21:33 < bridge> https://github.com/ddnet/ddnet/pull/7859 21:33 < bridge> but surprises me 21:33 < bridge> bcs `m_RenderingPaused = false;` is only called on a specfic sdl event 21:33 < bridge> mh ok 21:35 < bridge> who of you broke aiming 😫 21:35 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1200539558685720757/2024-01-26_21-34-30.mp4?ex=65c68c97&is=65b41797&hm=1abab3ec7996b061911ce928f51cd138a7c03ae044c5d068fbcf35d3ae4a1fde& 21:37 < bridge> happens on latest version 21:37 < bridge> ? 21:37 < bridge> why is the hookline so wrong 21:37 < bridge> it was never wrong for me 21:38 < bridge> r u all compiling with Ofast or smth 21:38 < bridge> just pulled from ddnet-git arch aur 21:38 < bridge> just pulled from ddnet-git aur 21:52 < bridge> if it also happens on a nightly from website 21:52 < bridge> report to github 21:53 < bridge> what website xd 21:53 < bridge> https://ddnet.org/news/nightly-build-revived/ 21:54 < bridge> Probably me 🥴 21:55 < bridge> yep 21:55 < bridge> its the smae 21:55 < bridge> its the same 21:56 < bridge> :banhammer: 21:57 < bridge> be quick and blame chiller 21:57 < bridge> https://github.com/jayphelps/git-blame-someone-else 22:03 < bridge> what did you change? 22:09 < bridge> be quick, blame chiller 22:26 < bridge> 22:26 < bridge> lol 22:27 < bridge> no one noticed anything since it works on default zoom 23:12 < bridge> it's about how using a tab width of 3 is better than 2 or 4 23:13 < bridge> he did another equally jarring talk about using lizard to ensure every single function in their codebase was below like 20 loc 23:13 < bridge> and he claims it cleaned it up a lot 23:18 < bridge> ^ @teero777 23:22 < bridge> it was always like that when cursor is close to tee 23:22 < bridge> @jupeyy_keks https://github.com/oasislinux/oasis 23:28 < bridge> No xd. The aimline was always correct 23:28 < bridge> when exactly on the tee, it was always wrong 23:28 < bridge> for close, but not exactly on, i don't remember 23:29 < bridge> Are u still stuck in 0.5? This is the future