00:10 < bridge> Full-time work on Github is probably cheating? 00:10 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1179922434833535016/Screenshot_2023-12-01_at_00.09.47.png?ex=657b8b66&is=65691666&hm=08f941e5dccaf8caf431c01afce260f54d09ac7a5b354c04955680be3fb94127& 00:16 < ChillerDragon> Is it grap share time? 00:16 < ChillerDragon> https://zillyhuhn.com/cs/.1701386257.png 00:18 < bridge> whats in your guy's opinion the best way to parse json stuff into client? - never did it, wanna do it. 00:18 < ChillerDragon> in the ddnet client? 00:18 < ChillerDragon> there is a json parser already 00:18 < bridge> y 00:18 < ChillerDragon> look at the server browser code on how its used 00:18 < bridge> aight - ty 02:42 < bridge> man it's already 1st December 04:20 < bridge> I'm implementing some commands, how do I keep them stored in settings_ddnet.cfg? 08:08 < bridge> 10+ / day on average is crazy 08:08 < bridge> very nice grid 😁 09:20 < bridge> FFR 09:35 < bridge> @jupeyy_keks do aoc 09:35 < bridge> https://adventofcode.com/ 09:35 < bridge> i got such a beautiful answer of first in rust 09:35 < bridge> but damn doing this in asm will be hard 09:36 < bridge> i need a hashmap xddd 09:36 < bridge> it has so much text, i'm too lazy to read it all 09:41 < bridge> ok gpt solved the puzzle. so not worth my time 09:41 < bridge> 😏 09:44 < bridge> boring 09:45 < bridge> well i need a C guy or girl to do it 09:45 < bridge> so i can beat it 09:46 < bridge> @jupeyy_keks for ffr u can read my blog 09:46 < bridge> 😬 09:46 < bridge> joking ill look later its gym time 10:07 < bridge> gm 10:45 < bridge> today i had a reverse deja vu again: 10:45 < bridge> I thought i've done exactly the same before but came to a different conclusion in the end 11:02 < bridge> xd 11:54 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1180099535930920970/image.png?ex=657c3056&is=6569bb56&hm=7291c83854ad470226e7683bc5fb5505416ae3d239852e52e86de07cf1a67cf4& 11:54 < bridge> @deen does ur work squash commits 11:54 < bridge> i dont remember if squashed count as 1 here 11:55 < bridge> chiller probs just does lot of small commits xd 11:55 < bridge> mine squashes 11:57 < bridge> I can squash, but usually don't. Each commit should make sense in isolation 12:03 < bridge> ic 12:03 < bridge> i think i follow more the philosophy of, a pr itself should make sense in isolatio, so inside it i make progress commits 12:03 < bridge> then they get squashed to the final clean version 12:03 < bridge> or rebased 12:04 < bridge> ofc some stuff is to complex 12:04 < bridge> omg a new firefox update 12:04 < bridge> ill do it tomorrow xd 12:04 < bridge> I follow a repo should make sense in isolation 12:04 < bridge> u mean a pr 12:05 < bridge> The full repo 12:05 < bridge> xddd 12:10 < bridge> My shower me was super clever today and noticed my current concept is really too overcomplicated and bloated. Now I'm super demotivated to fix it 12:11 < bridge> @ryozuki you have to do some stuff for ffr xd 12:11 < bridge> I'm out of ideas 12:15 < bridge> xd 12:16 < bridge> https://without.boats/blog/three-problems-of-pinning/ 12:17 < bridge> https://mcyoung.xyz/2023/11/27/simd-base64/ 12:25 < bridge> https://blog.marcocantu.com/blog/2023-november-turbopascal40.html 12:25 < bridge> xd 12:25 < bridge> Turbo Pascal was introduced by Borland in November 1983. It's officially turning 40 years old this month. 12:27 < bridge> I mean this is kinda rust related considered rust took quite a few concepts from it 😉 12:29 < bridge> Gpt truly ruined this industry forever 12:30 < bridge> I'll switch professions and become a baker 12:31 < bridge> Just like the baker will be useless, the coder will be. So you are free to do what you like 12:33 < bridge> Chatgpt is awful at anything food related 12:33 < bridge> Atleast for now maybe someone already has foodgpt in training 😄 12:34 < bridge> This is annoying with `git bisect`, have to use `--first-parent` for that 12:34 < bridge> Ask it what chemicals you need for making a mix that appeals to the human taste xd 12:35 < bridge> My main insight about AoC today was how to make a shebang for a SQL file 😄 https://github.com/def-/adventofcode-2023 12:35 < bridge> Well ai is ofc also trained for real motions at some point 12:35 < bridge> It fails miserably at trying to make anything balanced. It'll get the ingredients right but the ratios and portions will be all over the place 12:35 < bridge> oh lol 12:35 < bridge> 2kg of sugar lesgo 12:36 < bridge> General purpose robots taking over baking is the last thing I'd be worried about. Robotics is by far the jankest of all tech 12:36 < bridge> true 12:37 < bridge> Software cooler :kek: 12:38 < bridge> But you cant flex to normies with it 12:38 < bridge> Used to be. Now everyone is just using gpt for everything 12:39 < bridge> I dunno. I didn't underestimate it. Nature is basically nothing else than robotic+ ai 12:39 < bridge> Wouldn't 12:40 < bridge> Natures way of locomotion has nothing to do with the robotics we currently have access to. Robots have like 1 point in dex while humans have 100 12:41 < bridge> Special purpose robots will definitely take over most jobs but those don't need any ai sauce 12:41 < bridge> Yes but is that really a point against robotics? 12:41 < bridge> It doesn't have the disadvantage of evolution 12:42 < bridge> Yes, the strict limits on their motion is a huge issue. Muscles are far more flexible than servos 12:42 < bridge> That's not directly the point. A kitchen ai must still learn to act differently on different scenarios and normal programming simply is bad at this 12:43 < bridge> Making food in mass production ofc already happens 12:43 < bridge> The cutting edge general purpose robots have 1/10th of a humans output folding a shirt e.g. 12:44 < bridge> I mean that it's not happening tomorrow is clear. But let's say the next 10 years 12:44 < bridge> If you really want to shoehorn ai into it ai orchestrating a bunch of special purpose robots would probably be the goal unless a significant leap in robotics is upon us that I completely missed 12:46 < bridge> Sadly these hard coded robots are often not very flexible and thus only useful for very specific designed tasks.. thats nothing bad, and often probably more efficient. 12:46 < bridge> 12:46 < bridge> But the robot that will do your house works will not be this i guess^^ 12:47 < bridge> So i guess the interest in this general purpose designs will sky rocket this century 12:47 < bridge> I don't expect any robot to take over my housework completely for atleast 2 decades. Honestly they are faaar too clumsy and slow 12:47 < bridge> And decade 12:48 < bridge> Though yeah it only takes a handful of breakthroughs for that to change 12:48 < bridge> We still have 5 years 12:48 < bridge> Until Detroit become human takes place 12:48 < bridge> Xdd 12:49 < bridge> Sag wait 15 years 12:49 < bridge> 2038 12:49 < bridge> Ah wait 15 years 12:49 < bridge> You seem far too excited for what will probably be a catastrophic event for the working class 😄 12:50 < bridge> It's the only solution against human slavery 12:50 < bridge> i don't see a better one 12:50 < bridge> But yeah. I'm team machine 12:50 < bridge> If you think the people with the money will let us live in some startrek utopia you will be very surprised 12:50 < bridge> Humans have the disadvantage of being a nature being 12:51 < bridge> Too much hate inside us 12:51 < bridge> Hard to predict 12:52 < bridge> Not extremely hard to extrapolate from how the elite has acted in the last 2 millennia 12:53 < bridge> That's true, but if the ai is more intelligent than a human, we don't know what happens anyway.. why should it not be able to overthink whatever it was trained with 12:54 < bridge> They'll live in supercities with their ai robots and we'll live in slums and be used as court jesters 12:54 < bridge> Possible 12:54 < bridge> Maybe you will be one of them 12:54 < bridge> And change it 12:55 < bridge> Not everyone is a psychopath like musk and consorts 12:55 < bridge> Probably. So I'd rather the ai revolution chill for a little bit so I don't have to wear one of those jester outfits 12:56 < bridge> Oh you meant I'd be one of the megarich? I'd just blast myself off to mars so I don't die in the inevitable civil war 12:56 < bridge> Yes why not.. just gotta marry rich girl xdd 12:57 < bridge> I feel like it's more likely the poor ones will simply be sent to Mars 😂 12:57 < bridge> Who wants to be there xdd 12:57 < bridge> Ooor we can chill with all the AI and I can live a normal life then you can make all the cool robots after I die 12:58 < bridge> That's actually viable. Just ship all the poors of to some mining camp on another planet. Not that we'd need the ore, just so they are out of the way and have work 12:59 < bridge> Without ai you will lose 50% of your remaining awake life time to your job 12:59 < bridge> @learath2 when aoc 12:59 < bridge> C'est la vie 13:00 < bridge> Well c'est la social contract under capitalism to be more exact 13:00 < bridge> Xd 13:00 < bridge> nice 13:00 < bridge> Oh maybe this year I can follow it to the end. I don't have anything to do 13:01 < bridge> go 13:01 < bridge> do it in C 13:01 < bridge> share it 13:01 < bridge> And I wanted to practice my algs a bit 13:01 < bridge> i bench 13:01 < bridge> I was thinking Python this year 13:01 < bridge> :c 13:01 < bridge> I can do C too sure 13:01 < bridge> does c has btrees in std 13:02 < bridge> the program should load the input from a file 13:02 < bridge> called input.txt 13:02 < bridge> :kekW: 13:02 < bridge> xd 13:03 < bridge> You want btree, you make btree 13:03 < bridge> Honestly, not much of a point benching what I'd make against properly optimized data structures 13:04 < bridge> true 13:04 < bridge> Most proper libraries use hand vectorized code. I can't compete with that every day on every task 😄 13:04 < bridge> i was going to also use a optimized allocator and hashing algo 13:04 < bridge> tryharding 13:04 < bridge> @learath2 how hard is it to use mimalloc on C 13:04 < bridge> well just thinking about linking and stuff makes me already pain 13:05 < bridge> It should just be linking it in and it works 13:05 < bridge> I bet you can even just ld preload it like with jemalloc 13:07 < bridge> You don't need python. Gpt is really good in it xddd 13:07 < bridge> Ok sry i quickly rq 13:07 < bridge> ❯ hyperfine -N -w 5 ./target/release/rust-aoc 13:07 < bridge> Benchmark 1: ./target/release/rust-aoc 13:07 < bridge> Time (mean ± σ): 1.2 ms ± 0.1 ms [User: 1.0 ms, System: 0.2 ms] 13:07 < bridge> Range (min … max): 1.1 ms … 1.7 ms 2654 runs 13:07 < bridge> hmm 13:08 < bridge> 1.1ms sounds acceptable for like anything 😄 13:08 < bridge> Nice are we doing a competition about speed? 13:08 < bridge> yes 13:08 < bridge> I wonder if gpt can beat ryo 13:08 < bridge> this is not a optimized version 13:08 < bridge> its my first naive solution 13:08 < bridge> but i guess its p optimized 13:09 < bridge> im using a btreemap 13:09 < bridge> let me put everything in proper iterators 13:09 < bridge> i wanna see if it changes 13:09 < bridge> All the optimization juice is in here 13:09 < bridge> Atleast all the significant part 13:10 < bridge> ```py 13:10 < bridge> 13:10 < bridge> def sum_calibration_values(calibration_document): 13:10 < bridge> total_sum = 0 13:10 < bridge> for line in calibration_document: 13:10 < bridge> digits = [char for char in line if char.isdigit()] 13:10 < bridge> if digits: 13:10 < bridge> first_digit = digits[0] 13:10 < bridge> last_digit = digits[-1] 13:11 < bridge> two_digit_number = int(first_digit + last_digit) 13:11 < bridge> total_sum += two_digit_number 13:11 < bridge> return total_sum 13:11 < bridge> 13:11 < bridge> # Example usage: 13:11 < bridge> calibration_document = [ 13:11 < bridge> "1abc2", 13:11 < bridge> "pqr3stu8vwx", 13:11 < bridge> "a1b2c3d4e5f", 13:11 < bridge> "treb7uchet" 13:11 < bridge> ] 13:11 < bridge> 13:11 < bridge> result = sum_calibration_values(calibration_document) 13:11 < bridge> print("Total sum of calibration values:", result) 13:11 < bridge> ``` 13:11 < bridge> this was from gpt 13:11 < bridge> pls make sure it loads a file called input.txt 13:11 < bridge> i dont even know if it works for non example input xD 13:11 < bridge> and only outputs the result number 13:12 < bridge> do i look like a pyson dev. 13:12 < bridge> gpt is down for me, cant do it 13:13 < bridge> I can't wait for the future where the company just sends everyone home when gpt goes down because no one can code without it anymore 13:13 < bridge> same 13:13 < bridge> and at home i do the real fun stuff 13:14 < bridge> what awesome times that will be 13:15 < bridge> Ur ofc not getting paid for those hours 13:15 < bridge> Imagine thinking capitalists will pay you for nonproductive hours :xDe: 13:16 < bridge> i dont think you understand how that works 13:16 < bridge> if you have a skill nobody has. you can do whatever you want. because you are basically their golden shovel 13:16 < bridge> without you they can't even dig 13:17 < bridge> if you have a job that is very common, then u the real slave 13:17 < bridge> that's where capitalism shows it's evilness 13:17 < bridge> my iterator version is slower :o 13:17 < bridge> || ```rust 13:17 < bridge> let result = input.lines().fold(0u32, |acc, line| { 13:17 < bridge> let indices: BTreeMap<_, _> = numbers 13:17 < bridge> .keys() 13:17 < bridge> .flat_map(|x| line.match_indices(*x)) 13:17 < bridge> .collect(); 13:17 < bridge> 13:17 < bridge> let first = numbers.get(indices.first_key_value().unwrap().1).unwrap(); 13:17 < bridge> let last = numbers.get(indices.last_key_value().unwrap().1).unwrap(); 13:17 < bridge> 13:17 < bridge> acc + first * 10 + last 13:17 < bridge> }); 13:17 < bridge> ``` || 13:18 < bridge> A skill in prompting chatgpt is much more replaceable than actually being able to code 13:18 < bridge> || ```rust 13:18 < bridge> let mut result = 0; 13:18 < bridge> let mut indices = BTreeMap::new(); 13:18 < bridge> for line in input.lines() { 13:18 < bridge> for x in numbers.keys() { 13:18 < bridge> indices.extend(line.match_indices(x)); 13:18 < bridge> } 13:18 < bridge> 13:18 < bridge> let first = numbers.get(indices.first_key_value().unwrap().1).unwrap(); 13:18 < bridge> let last = numbers.get(indices.last_key_value().unwrap().1).unwrap(); 13:18 < bridge> 13:18 < bridge> result += first * 10 + last; 13:18 < bridge> indices.clear(); 13:18 < bridge> } 13:18 < bridge> ``` || 13:18 < bridge> this is faster 13:18 < bridge> 1.2ms vs 1.5ms 13:18 < bridge> unsafe on the unwrap changes nothing 13:19 < bridge> Have you run it a couple more times? 13:19 < bridge> correct, but the fact you still have to think is already smth most can't do 😂 13:19 < bridge> hyperfine does that 13:19 < bridge> it ran it 2700 times 13:19 < bridge> Benchmark 1: ./target/release/rust-aoc 13:19 < bridge> Time (mean ± σ): 1.5 ms ± 0.1 ms [User: 1.2 ms, System: 0.2 ms] 13:19 < bridge> Range (min … max): 1.3 ms … 1.9 ms 2245 runs 13:19 < bridge> iterator 13:19 < bridge> Benchmark 1: ./target/release/rust-aoc 13:19 < bridge> Time (mean ± σ): 1.2 ms ± 0.1 ms [User: 1.0 ms, System: 0.2 ms] 13:19 < bridge> Range (min … max): 1.1 ms … 2.6 ms 2706 runs 13:19 < bridge> for loop 13:20 < bridge> Curious, maybe check the assembly generated? 13:20 < bridge> can u send me the input file? without account i cant seem to access it 13:21 < bridge> i make you the fastest ever 13:21 < bridge> but ur computer specs is diferent 13:21 < bridge> i give u the code 13:21 < bridge> https://gist.github.com/edg-l/5c65631ed4d9e736187058a955f97556 13:21 < bridge> The fastest ever is just puts the result 13:21 < bridge> result is 54277 13:22 < bridge> (note it does nto work for others cuz each gets their own input) 13:22 < bridge> `puts("54277")` beat this one nerd 13:22 < bridge> xd 13:22 < bridge> why call puts use a syscall 13:23 < bridge> Portable 😄 13:23 < bridge> using ahash on the numbers hashmap doesnt help 13:24 < bridge> @ryozuki what is the assignment to numbers in your example? 13:24 < bridge> you didnt send full code xd 13:25 < bridge> ``` 13:25 < bridge> let mut numbers = HashMap::with_capacity(18); 13:25 < bridge> numbers.insert("one", 1); 13:25 < bridge> numbers.insert("two", 2); 13:25 < bridge> numbers.insert("three", 3); 13:25 < bridge> numbers.insert("four", 4); 13:25 < bridge> numbers.insert("five", 5); 13:25 < bridge> numbers.insert("six", 6); 13:25 < bridge> numbers.insert("seven", 7); 13:25 < bridge> numbers.insert("eight", 8); 13:25 < bridge> numbers.insert("nine", 9); 13:25 < bridge> numbers.insert("1", 1); 13:25 < bridge> numbers.insert("2", 2); 13:25 < bridge> numbers.insert("3", 3); 13:25 < bridge> numbers.insert("4", 4); 13:25 < bridge> numbers.insert("5", 5); 13:25 < bridge> numbers.insert("6", 6); 13:25 < bridge> numbers.insert("7", 7); 13:25 < bridge> numbers.insert("8", 8); 13:25 < bridge> numbers.insert("9", 9); 13:25 < bridge> ``` 13:25 < bridge> xd 13:25 < bridge> || ```rust 13:25 < bridge> use std::{ 13:25 < bridge> collections::{BTreeMap, HashMap}, 13:25 < bridge> error::Error, 13:25 < bridge> }; 13:25 < bridge> 13:25 < bridge> use mimalloc::MiMalloc; 13:25 < bridge> 13:25 < bridge> #[global_allocator] 13:25 < bridge> static GLOBAL: MiMalloc = MiMalloc; 13:25 < bridge> 13:26 < bridge> fn main() -> Result<(), Box> { 13:26 < bridge> let input = std::fs::read_to_string("input.txt")?; 13:26 < bridge> 13:26 < bridge> let mut numbers = HashMap::with_capacity(18); 13:26 < bridge> numbers.insert("one", 1); 13:26 < bridge> numbers.insert("two", 2); 13:26 < bridge> numbers.insert("three", 3); 13:26 < bridge> numbers.insert("four", 4); 13:26 < bridge> numbers.insert("five", 5); 13:26 < bridge> numbers.insert("six", 6); 13:26 < bridge> numbers.insert("seven", 7); 13:26 < bridge> numbers.insert("eight", 8); 13:26 < bridge> numbers.insert("nine", 9); 13:26 < bridge> numbers.insert("1", 1); 13:26 < bridge> numbers.insert("2", 2); 13:26 < bridge> numbers.insert("3", 3); 13:26 < bridge> numbers.insert("4", 4); 13:26 < bridge> numbers.insert("5", 5); 13:26 < bridge> numbers.insert("6", 6); 13:26 < bridge> numbers.insert("7", 7); 13:26 < bridge> full code 13:26 < bridge> lmao 13:26 < bridge> oh wait 13:26 < bridge> rayon 13:26 < bridge> its threads time 13:27 < bridge> Did you find out why slower? 13:28 < bridge> This task is so quick that I would think spinning up a thread and synchronization overhead might make it worse 13:29 < bridge> one thing i noted is 13:29 < bridge> im doing the sum in the fold body 13:29 < bridge> while i can just return the result 13:29 < bridge> and do .sum 13:29 < bridge> ``` 13:29 < bridge> let result: u32 = input.lines().par_bridge().fold(|| 0u32, |acc, line| { 13:29 < bridge> let indices: BTreeMap<_, _> = numbers 13:29 < bridge> .keys() 13:29 < bridge> .flat_map(|x| line.match_indices(*x)) 13:29 < bridge> .collect(); 13:29 < bridge> 13:29 < bridge> let first = numbers.get(indices.first_key_value().unwrap().1).unwrap(); 13:29 < bridge> let last = numbers.get(indices.last_key_value().unwrap().1).unwrap(); 13:29 < bridge> 13:29 < bridge> first * 10 + last 13:29 < bridge> }).sum(); 13:29 < bridge> ``` 13:29 < bridge> this is faster 13:29 < bridge> rayon 13:30 < bridge> 1.1ms 13:30 < bridge> wait 13:30 < bridge> i dont need fold at all 13:30 < bridge> ```rust 13:30 < bridge> let result: u32 = input.lines().par_bridge().map(|line| { 13:30 < bridge> let indices: BTreeMap<_, _> = numbers 13:30 < bridge> .keys() 13:30 < bridge> .flat_map(|x| line.match_indices(*x)) 13:30 < bridge> .collect(); 13:30 < bridge> 13:30 < bridge> let first = numbers.get(indices.first_key_value().unwrap().1).unwrap(); 13:30 < bridge> let last = numbers.get(indices.last_key_value().unwrap().1).unwrap(); 13:30 < bridge> 13:30 < bridge> first * 10 + last 13:30 < bridge> }).sum(); 13:30 < bridge> ``` 13:30 < bridge> same speed 13:30 < bridge> xd 13:30 < bridge> 1.1 13:30 < bridge> without rayon 1.5ms 13:31 < bridge> Wasnt your attempt without iterators also 1.1? 13:31 < bridge> 1.2 13:31 < bridge> lol was about to 14:09 < bridge> ok i took your solution and tried to remove the btree: 14:09 < bridge> 14:09 < bridge> ```rs 14:09 < bridge> let result: i32 = input 14:09 < bridge> .par_lines() 14:09 < bridge> .map(|line| { 14:09 < bridge> let matches = numbers.keys().map(|x| line.match_indices(x)); 14:09 < bridge> let m = matches.flatten().peekable(); 14:09 < bridge> 14:09 < bridge> let (min, max) = match m.minmax_by_key(|x| x.0) { 14:09 < bridge> MinMaxResult::NoElements => todo!(), 14:09 < bridge> MinMaxResult::OneElement(min) => (min, min), 14:09 < bridge> MinMaxResult::MinMax(min, max) => (min, max), 14:09 < bridge> }; 14:09 < bridge> 14:09 < bridge> let first = numbers.get(min.1).unwrap(); 14:09 < bridge> let last = numbers.get(max.1).unwrap(); 14:09 < bridge> 14:09 < bridge> first * 10 + last 14:09 < bridge> }) 14:09 < bridge> .sum(); 14:09 < bridge> ``` 14:09 < bridge> 14:09 < bridge> i wonder if that makes any difference at all 14:09 < bridge> i also used itertools 14:09 < bridge> ah 14:10 < bridge> oh 14:10 < bridge>