Advent of Code 2022

[trail hemlock]

The Advent of Code 2022 is happening in just a few days!
https://adventofcode.com

@red pelican created a private leaderboard for this server.
Private advent of code 2022 leaderboard code:

2265237-0f8ae43d

You can join if you log in here https://adventofcode.com/2022/leaderboard and click the private leaderboard joining option!

Tell everyone in what language you're gonna do it! Personally I think I'm still gonna go with Rust, like I did in 2020 :)
Or maybe you have a different gimmick, like focusing on some kind of pretty visualisation of the problem after solving it ? 👀

[nova mango]

I'm gonna go with Kotlin, I think it has a lot of things that would be useful here

[mental owl]

i'll try to do it with rust i think

this time around there are no exams to distract me from it

[tardy barn]

I'm doing last year's with Rust, it's so much fun

[red pelican]

I havent done advent of code before, 100% doing it with rust

might be partially busy with school work but I know im going to have some time for it regardless

[tardy barn]

^^

[modern delta]

I was thinking of doing it for the first time, and use rust as a training exercise !

[red pelican]

aw this is in the advent of code HTML head ```Oh, hello! Funny seeing you here.

I appreciate your enthusiasm, but you aren't going to find much down here.
There certainly aren't clues to any of the puzzles. The best surprises don't
even appear in the source until you unlock them for real.

Please be careful with automated requests; I'm not a massive company, and I can
only take so much traffic. Please be considerate so that everyone gets to play.

If you're curious about how Advent of Code works, it's running on some custom
Perl code. Other than a few integrations (auth, analytics, social media), I
built the whole thing myself, including the design, animations, prose, and all
of the puzzles.

The puzzles are most of the work; preparing a new calendar and a new set of
puzzles each year takes all of my free time for 4-5 months. A lot of effort
went into building this thing - I hope you're enjoying playing it as much as I
enjoyed making it for you!

If you'd like to hang out, I'm @ericwastl on Twitter.

• Eric Wastl```

[trail hemlock]

yeah :D

[tardy barn]

That's so cool
This whole event is so nice, I'm glad it exists

[near steppe]

I do AoC in javascript typically because I'm strong with the lang, and its dynamic nature helps with quick typing since I try to compete fr. This year I resolve to do a Rust rewrite each day, so I can practice rust without having to worry about understanding the problem at the same time

[delicate basalt]

i'll prob do it in python since i'm not really familiar with rust

[trail hemlock]

Why not take this opportunity to get more familiar with it? 👀

[delicate basalt]

too scared to leave my comfort zone

[trail hemlock]

Why do the advent of code then? :(

(I'm kidding, you don't need to get out of your comfort zone, but still, it's a great opportunity to do so)

Ok better phrasing: why are you scared to leave your comfort zone?

[delicate basalt]

not sure

[tardy barn]

Who isn't?

[trail hemlock]

I guess you've got a point :v

but like idk, I'd try if that's what I wanted to do at some point, AoC would be a perfect opportunity xd

[mental owl]

@amber siren 🐝

[amber siren]

[pt]

[marble stream]

o/

[red pelican]

private leaderboard code is in the initial message

[marble stream]

ok

im just doing the 2015 one

cos ill probably just do 2 or 3 puzzles then give up

[marble stream]

ive done the first 4 days of 2015 in c

[marble stream]

im stuck on 2015 5.2

lol

[marble stream]

:w

om

omg

[tardy barn]

Mfw trying to exit discord

[mental owl]

😭

it won't let us leave

[marble stream]

lmao

i need to stop wroking on this puzzle because

im getting annoyed LOL

[tranquil ether]

Anyone looking to do their AoC in Rust should consider looking at my template on GitHub: https://github.com/MitchellMarinoDev/aoc_template_rs.

[trail hemlock]

daaaaamn

[tardy barn]

Nice!

[mental owl]

i wonder if there'll be someone doing aoc in brainfuck

[marble stream]

I think it could be fun for like

The first day

[mental owl]

yeah

[marble stream]

I am a terrible programmer I am just on the sixth day of 2015 in C

[mental owl]

https://cdn.discordapp.com/attachments/1007560833225732106/1029854767381487686/unknown.png

when this aoc starts we can horribly fail in parallel

[marble stream]

Lol ok

I love C Tbh

U just let it RIP

[livid pier]

I share the same feeling TBH

I want to do it in Rust, but I kinda just want to do it quick and dirty in python like I usually do, then make it pretty in Rust

1-25 December is python, then 1-25 January is Rust 😄

I will venture to do it in Rust for as long as I can reasonable abide, if it feels dreadful then I might switch over midway

I've already started doing some re-writes for past years and have updated my repo.

So I have a nice filestructure already ready for it if I feel the need to switch back and forth but I'm going to give it an honest attempt at Rust from the start

[near steppe]

1-25 December is python, then 1-25 January is Rust 😄
yeah I'm feeling that too, but I know I won't end up doing 1-25 January in Rust :(

[livid pier]

OH man it starts at midnight US East. Tragic.

[tardy barn]

Starts <t:1669870740:R>!

[mental owl]

it's here

[modern delta]

Oh no it’s December already

[mental owl]

yeah this terrible year is coming to an end

[crude juniper]

I'm giving it a go in Rust! but will fall back to Python if my inexperience makes a particular day take too long

[red pelican]

time to do the first task

[mental owl]

mu first task went like this so far
#newbie-advice message

i want to do aoc in something that isn't my daily comfort language but life is fucking me over the second i want to step out :(

[red pelican]

both parts done

im also practicing/learning vim at the same time which is fun

im getting access to the input like thisrs static INPUT: &str = include_str!("input.txt");

[mental owl]

fuck it i'm doing it in c# unless/until i get other option fixed

better than not doing it

[nova mango]

I'm building a tool that automatically runs the code when a file changes and checks against the examples provided

with Kotlin

[tardy barn]

Help
I think I'm fundamentally misunderstanding the puzzle
(Part 2)

Neither one works

[mental owl]

that seems...wrong

link to aoc spoilers ||If anyone cares: https://gitlab.com/tlppi/aoc-2022/-/tree/master/one||

[red pelican]

a tip with only one word, but should make it much easier: ||sort||

[trail hemlock]

aight time to start

[tardy barn]

Oh

Huh

That's so much easier

I don't understand why my very-backwards-and-dumb method doesn't work though

[mental owl]

Did some pretty fun refactoring https://gitlab.com/tlppi/aoc-2022/-/commit/d28de0dace2ea06592049005efb641924031b1c5

iters are good shit

[zealous tulip]

Sigh I might do it in rust even though I have mock exams

[zealous tulip]

I'm pretty happy with day one

[livid pier]

This was my solution in Rust

My friend made fun of me because I added error handling. 😦

[tardy barn]

Imagine doing error handling
Hah
bullies

[zealous tulip]

you could use a unwrap_or_default() or a or(Ok(0)) if you wanted to handle the parsing error

[livid pier]

Yeah I was going to do unwrap_or(0) or something like that, but I wanted to know WHEN an error happened as well rather than just kind of ignoring it. Which showed me that ||it didn't like the end of file area which was what lead me to adding the trim() on the initial split||

I appreciate the recommendation though!

[zealous tulip]

tru

||ngl I thought the end of the file was my copying and pasting incorrectly so I removed it 😬 ||

[zealous tulip]

||although that can be fixed by using split_terminator instead of just split||

[livid pier]

||I wasn't sure if it was or wasn't but I thought "hey bit of serendipity, I now have an edgecase I have fixed"||

Highly respect your solution to that problem though

✍️ "look up split_terminator"

[zealous tulip]

Idk if my solution for the 2nd is that good of a idea due the fact I iterate a lot

I'd benchmark mark it versus ||vec.sort();vec.reverse();|| but eh

[livid pier]

I'd like to see the results.

[zealous tulip]

fine 🙄

[zealous tulip]

(I will be assuming their input is going to be a iterator which means I'll be counting the vecs collect)

@livid pier it is only slightly faster

[livid pier]

Hahaha only a little bit

[zealous tulip]

though this gives me a idea...

||the idea was terrible (I was sorting the top_3 vector), .sort is REALLY slow huh||

[zealous tulip]

I did not notice that it was actually a whole unit different

[livid pier]

I was goign to say

[zealous tulip]

that's really big, huh

[livid pier]

What are you using for benchmarking

[zealous tulip]

criterion

very good

[livid pier]

I figured

[red pelican]

you can bring the input text into the program at compile time with include_str!("input.txt")

[zealous tulip]

trueee

very good point

though that won't help me much since I want to replace the new lines

those were so annoying

technically for solving this EXACT input \r\n works

but i'd rather not

[zealous tulip]

actually it makes it so I can put the input under day-1\src\input.txt

which is handy

[livid pier]

Going to do the same with my code. Current implementation and the first "just do it as fast as possible" and see if i did optimise it or if I just did the same thing but with more complex looking code

[zealous tulip]

If I had set up my stuff correctly I could have had criterion look at the change in performance over time as I try to improve a function

Hrm maybe I should use lai 🤔 https://bheisler.github.io/criterion.rs/book/iai/comparison.html

[livid pier]

Its insane how much the prints take.

I ran a bench with the println! and got 207.18 µs
I ran a bench with removing and got 83.829 µs

[zealous tulip]

it does a lot!

and we are talking microseconds here

[zealous tulip]

since it also has to format often

[livid pier]

Yeah and for a thing I will run one time a year this is a lot!

[zealous tulip]

true

I wonder why mine is so much higher

I am not printing anything

[livid pier]

OH i thought your bench here was your whole pogram

[zealous tulip]

oops

[livid pier]

I was like oh man you did yours in 530 nano seconds 😢

[zealous tulip]

I was benching these two things

[livid pier]

Ahhhhh

I removed the reverse (and just took the last 3 values at the end of the vector) and it took 82.579 µs down to 73.326 µs

[zealous tulip]

oh yeah

[livid pier]

So the reverse was worth about 9 microseconds

[zealous tulip]

let me bench that rq

517.98 ns vs 5.8426 µs

hrm

maybe if I pass in a vector directly

actually let me test these two approaches in the scope of the whole thing

[livid pier]

Proud of that. Final implementation without the reverse vs the intial implementation (no prints for either)

[zealous tulip]

can you show source code?

now that I tweaked some things I am getting the result that sorting the vec is better

I am doing this

[mental owl]

what are you doing this with?

(the benchmarking)

[zealous tulip]

that was benchmarking the day_1 code I just posted

[mental owl]

i mean, how are you benchmarking it?

[zealous tulip]

ah

[mental owl]

never bothered to benchmark before, was gonna see how mine was doing

[zealous tulip]

criterion

[mental owl]

can you share the setup?

im trying and feeling confused

or maybe i'm being stupid...

[zealous tulip]

you need to have a folder in the root of your project

then have

[[bench]]
name = "name of where your benchmarks will be inside /benches"
harness = false

wait nohrm

edited to fix

[mental owl]

yeah i got that bit

tbh what i can't figure out is how to access my crate

which seems stupid

[zealous tulip]

huh?

ohh

your crate needs to be a lib

well, I don't think so

but you can just import it with it's name

[mental owl]

ohh eww thats too much effort tbh

[zealous tulip]

I use a separate project to put all my

benchmarks in

[mental owl]

is there a way to get criterion to just benchmark my tests? cargo bench does but it's not exactly, uh, detailed information

[zealous tulip]

no :(

[mental owl]

annoying

meh, whateer

[zealous tulip]

can you give me your function

[mental owl]

is on the gitlab i posted earlier but its a bit of a mess because im trying to keep both the first and second parts around

[zealous tulip]

ah

I am benchmarking both of them

[mental owl]

😆 looks like println! isn't really the problem in mine

its just slow in general

[zealous tulip]

so that initial .replace is costing me 500 µs

[mental owl]

...
...
sed

[zealous tulip]

if only I was on linux

also does it really count if we are modifying the input

before testing

ignore how doing it at compile time technically also does that

[mental owl]

i was considering just changing the input to just be able to be include!ed by setting it up as an array

it didn't work but was a fun thought

[livid pier]

Deleted because I read it wrong

[zealous tulip]

for the sake of not modifying anything at compile time I'll just do it using windows new lines

[livid pier]

Thought you could spoiler code

[zealous tulip]

nope

weirdly

[livid pier]

Thats my final implementation with no reverse and prints included

[zealous tulip]

these are my final 2 with the performance improvement being from handling windows lines directly

to note the lines is slower then calling directly split_terminator("\r\n")

this is because what lines does under the hood is call split_terminator('\n') and then a map which removes all \r at the end of the function

this causes a 10 µs difference for me

[mental owl]

is there a non-mutating sort() in std?

(as in, so that it can be chained)

[zealous tulip]

As far as I know, no

[mental owl]

or is there a way to sort a const vec

[zealous tulip]

doing things in constant is so frustrating

it's what I imagine working without standard feels like but probably much better in reality

Upon further realisation I think the reason why I went down this weird road for my code initially was because I thought that all my previous code for the first solution was pretty good and so did not think about how I could optimise both solutions together but instead just solution 1 and 2 separately

My first solution would be good if I wasn't measuring running both solutions at once

[livid pier]

This is my .split("\n") vs .lines()

[zealous tulip]

hmmm

[livid pier]

Ran it back to back like 7 times and it seems to be all around the 66 microseconds vs 69

[red pelican]

Not on iterators, but the itertools crate does which is a small and very widely used crate

[mental owl]

yeah I'm trying to not use external crates

[zealous tulip]

Time to use the aoc_2022 crate /j

[red pelican]

Why?

[mental owl]

fun

[zealous tulip]

i was doing to use urq but then I realised authentication for it existed

[mental owl]

i mean, the implementation is just

trait Sorted<T: Ord> {
   fn Sorted(self) -> Self
}

impl<T: Ord> Sorted<T> for Vec<T> {
   fn sorted(self) ->  Self {
       self.sort();
       self
   }
}

but yk

(written on phone might be wrong but my point stands)

[delicate basalt]

who's the one anon lol

i feel like being anon is pretty pointless because:
there's only 1
everyone on the board is likely following this thread

[crude juniper]

This was my go!

It's the first Rust program I've written from scratch (i.e. not a rustling) so any advice much appreciated 🙂

[zealous tulip]

Since a iterator needs to go through all the items in the list it would be a bit slow to get part1 and part2 with them, can you think with a faster way?

Especially since you are already sorting them

[crude juniper]

Ah! There's no need to iterate at all!

Thx 🙂

[zealous tulip]

Do you have to reverse it 🤔

[crude juniper]

I suppose not

Is there a quick way to do indexes-from-the-end?

[zealous tulip]

I don't think so

[crude juniper]

I'm used to Python having negative indexes, but it doesn't seem to be a thing in Rust

So I guess I just need to grab the length, and use that

fn main() -> anyhow::Result<()> {

    let mut elf_calories_sum: Vec<i64> = include_str!("calories.txt")
        .split("\n\n")
        .map(|s| 
            s.split('\n')
            .map(|y| y.parse::<i64>().unwrap())
            .sum()
        )
        .collect();
    elf_calories_sum.sort();

    let final_index = elf_calories_sum.len() - 1;
    let part1: i64 = elf_calories_sum[final_index];
    dbg!(part1);
    let part2: i64 = elf_calories_sum[final_index] + elf_calories_sum[final_index-1] + elf_calories_sum[final_index-2];
    dbg!(part2);
    
    Ok(())
}

Like that!

[zealous tulip]

Hnmm

For part 2 I used a iterator over a range instead of getting them one by one

But maybe this one is faster (it probably is)

I'd bench it if I wasn't in my bed

[crude juniper]

Like so?

let part2: i64 = elf_calories_sum[..3].iter().sum();

[zealous tulip]

Yep

[crude juniper]

Oh that's makes a different number!

I'm at the wrong end of the array

[zealous tulip]

Ah yes the range is wrong

Yep

It'd need to be the length - 3

And the length not the final index

[crude juniper]

let part2: i64 = elf_calories_sum[final_index-2..].iter().sum();

[zealous tulip]

Hmmm

[crude juniper]

That does the trick!

[zealous tulip]

While it's cleaner it probably has some performance downside

[crude juniper]

I'd definitely use the iter if they wanted many more than the first 3!

What do you use to benchmark?

[zealous tulip]

Criterion

I am getting out of bed for this I NEED to know

[zealous tulip]

Tru

[crude juniper]

I need to do a clothes wash before I hit the hay, but will try Criterion when back

[trail hemlock]

https://tenor.com/view/sheldon-cooper-bbt-sleep-i-dont-need-sleep-gif-5115394

[zealous tulip]

the possible TINIEST difference

I will rerun with a higher sample size

[crude juniper]

Wow that's close!

[zealous tulip]

newer results prove they are pretty much the same

[crude juniper]

That's good to know for the future

and Criterion looks cool! I'll be experimenting with that tomorrow

Also the getting_started page having a benchmark with values in picoseconds!! 0_o

[zealous tulip]

Well if the time was longer it'd need to be a more complicated example

[zealous tulip]

another rerun with me shutting off from the pc apart from vscode gave this

this is a sample size of 100 and a measurement time of 1 minute

I will now rest

[zealous tulip]

at minimum the difference is 0.038 microseconds and at max it's 0.371

[livid pier]

I also wanted to do this too. So I just did [vector.len()-3..] or [vector.len()-1]

[zealous tulip]

To me the final index variable is more intuitive

But it's not intuitive when you do it with ranges since it uses count not length

[mental owl]

I cheated in d1t2 by using .net Framework builtin types

[red pelican]

For only the last one you can do .last().unwrap()

[frank cairn]

my solution is currently

use std::fs;

fn main() {
    let f = fs::read_to_string("aocinput").unwrap();
    let mut elves = Vec::new();
    for elf in f.split("\n\n") {
        let mut r = Elf { food_items: Vec::new() };
        for food_item in elf.split("\n") {
            if let Ok(x) = food_item.parse() {
                r.food_items.push(x);
            }
        }
        elves.push(r);
    }
    println!("{}", answer(elves));
}

struct Elf {
    food_items: Vec<u32>,
}

fn answer(elves: Vec<Elf>) -> u32 {
    most_calories(elves)
}

fn most_calories(elves: Vec<Elf>) -> u32 {
    let mut amount = 0;
    let mut best_three = [(0usize, 0u32); 3];
    let mut max = u32::MAX;
    for best in best_three.iter_mut() {
        let mut current_best = (0usize, 0u32);
        for (i, elf) in elves.iter().enumerate() {
            let mut count = 0;
            for food_item in &elf.food_items {
                count += food_item;
            }
            if count > current_best.1 && count < max {
                current_best = (i, count);
            }
        }
        *best = current_best;
        max = current_best.1;
        amount += current_best.1;
    }
    amount
}

it runs in linear time

[trail hemlock]

...and so I still haven't started

[dapper abyss]

iam done

--------Part 1--------- --------Part 2---------
Day Time Rank Score Time Rank Score
1 >24h 134535 0 >24h 129276 0

took like an hour in steep2learn :rust:

[red pelican]

have you joined the private leaderboard? 2265237-0f8ae43d

[modern delta]

If anyone is interested this is my d01 solution :D https://github.com/eriizu/aoc_2022/blob/main/src/days/d01.rs

I used the template from https://github.com/MitchellMarinoDev/aoc_template_rs listed earlier

I also did the 1st part of d02 https://github.com/eriizu/aoc_2022/blob/main/src/days/d02.rs

I kinda want bigger inputs for day 2 so I can MULTITHREAD IT

[red pelican]

my day 2 solution. made it idiomatic and as easy to reason about as I could
||https://pastebin.com/P59CT4zC||

and used let else once 🙂

[livid pier]

Where do you see that?

Ah I see

--------Part 1-------- --------Part 2--------
Day Time Rank Score Time Rank Score
2 06:47:42 61256 0 06:59:11 56633 0
1 05:57:32 46160 0 06:18:57 45036 0

[livid pier]

Part 1: (Ugly)|| (This is the function to calculate the score for an individual line, the rest is just a loop and print lines to display the score)||

Part 2: || The same but have just updated the arms of the match statements ||

[livid pier]

Changed essentially 3 lines and got a 45% performance gain

||I moved the splitting string to a vector out of the functions and did it once in the loop and passed the vector as a parameter to the get_score functions||

[red pelican]

tip, you can match on a tuple like this ||rs match (match_info[1], match_info[0]) { ("X", "A") => ..., ... }||

[livid pier]

Now that's hot

[frank cairn]

should probably show my result

its very idiomatic as in it makes invalid states unrepresentable by using enums for everything

one moment

[red pelican]

same

[red pelican]

here

[frank cairn]

this is for part 2

[red pelican]

cool, you did a lot differently to me but both are good

[livid pier]

;compile

[undone creek]

Critical error:

You must provide a valid language or compiler!

;compile c++
```
int main() {}
```

Requested by: MattHasFun#9865

[frank cairn]

it reads a file

[livid pier]

Ah

[frank cairn]

;compile rs

[undone creek]

Critical error:

Unable to find compiler or language for target 'rs'.

Requested by: TudbuT#2624

[livid pier]

Does that code compile on your side?

[frank cairn]

;compile rust

[undone creek]

Critical error:

You must attach a code-block containing code to your message or quote a message that has one.

Requested by: TudbuT#2624

[frank cairn]

yes

[frank cairn]

;compile rust

[undone creek]

Critical error:

Unable to find compiler or language for target 'rs'.

Requested by: TudbuT#2624

[frank cairn]

Huh

just exchange the two score += lines

[red pelican]

#[derive(Clone, Copy)] should make it work

[frank cairn]

not needed

just exchange the two lines

[mental owl]

groans yk when it works on the test input but not on the actual one?

yea i had that too

(i was stupid and mixed up int values for enums)

but like... how do i /debug/

as a c# pleb i get the benefits of breakpoints

uwu

[red pelican]

Ive gotten all 4 correct on the first go so far by just making sure my logic is right

[zealous tulip]

day 2 https://github.com/Coca162/advent-of-code-2022/blob/main/day-2/src/main.rs
||For this one I used the phf crate which is just for lookup tables and I also decided I would be stupid and represent the values/scores of things with what the enum is stored as||

[frank cairn]

#1044250755953078312

[zealous tulip]

that is indeed the channel

[mental owl]

mmm yes the channel here is made out of channel

[frank cairn]

oh

i thought this was off topic because i am very stoopid

sorry

[mental owl]

https://media.discordapp.net/attachments/1022233702186623056/1040795534052560957/watermark.gif

[zealous tulip]

https://media.discordapp.net/attachments/941532927186067506/1030755052610125925/caption_16.gif

[frank cairn]

awwwwwww

[mental owl]

u irl

[frank cairn]

this boi is so cute

[mental owl]

if only I allowed myself access to external crates :(

(for part two)

probably could do the same for part one (by that I mean absolutely i could) but this was easier because I don't have to deal with two options

[zealous tulip]

they are being unwrapped either way

and_then is also a thing if you wanted to handle at once

[crude juniper]

Day 2 complete! Writing this really messed with my head to begin with, but I think got my Rust hat on in the end 🙂 https://gist.github.com/JELAshford/1cd79f8ef0d9bd74424e26ea6fc712ac

I'm not sure how I'd approach the problem if there were many more than 3 states for tool/game outcome each.

[red pelican]

looks alright.
id_to_tool should have just been a standalone function instead of an associated function of GameState imo.
do you know rustfmt? its the formatter that the entire rust ecosystem uses. through the terminal its cargo fmt, but you should be able to find a shortcut for it in whatever editor you use

[frail slate]

hi i joined

[crude juniper]

Just set up rustfmt on Save, looks a lot nicer now

and I agree about id_to_tool

[red pelican]

my day 3 solution for both parts ||https://pastebin.com/zd98VVpM 52 lines of code when not including unit tests.||
steadily getting better at vim, this is very good practice

hmm let me test how fast it is

around 270 microseconds with file reading and printing not included

oh 75 microseconds by just removing my use of ||hashset||

not worth it for sets this small

[frail slate]

consider using || hashet::intersection || lol

|| its kinda funny that you used sets and then didn't use them at all ||

[red pelican]

what about the second part?

[red pelican]

doing this made it significantly slower, 360 microseconds

[frail slate]

tbf i just || collected back into a hashset and did another intersection lmao ||

ok well idc about speed

[red pelican]

it seems to me that use of ||HashSet|| isnt anywhere near worth it for this small scale stuff

[frail slate]

fair enough

but basically all of the functions i needed were implemented already

[red pelican]

ok this is my solution with ||HashSet|| use removed and timer added ||https://pastebin.com/Phqsv2rX||

[zealous tulip]

day 3: ||https://github.com/Coca162/advent-of-code-2022/blob/main/day-3/src/main.rs WOW the 2nd star was really annoying I was think of using a hashset but then realised that I don't want the most used character between all 3 so I used a pure iter solution||

||Chunks was VERY useful for the 2nd one, though it requires I collect the lines into a vec :/||

[frail slate]

|| i used itertools so i didn't need to collect but that probably wasn't that beneficial anyway ||

[zealous tulip]

||damn that find in find_badge_item_type is much better then what I was thinking when I made mine||

[modern delta]

Wow! this is very elegant! I didn't know about chunks, I'll try and use it in the future.

[zealous tulip]

I have yet to use a for loop for advent of code

I think I will try to make all of them loopless

[modern delta]

That's a cool challenge

[mental owl]

yall doing some weird rust challenges adjacent to it and I'm just wondering when I'll drop dead with my c#

[modern delta]

my goal is only trying to complete the AOC in rust

I'm glad that it enables me to get some practice with day to day stuff I need to master

[mental owl]

did AoC with a splitting headache

look at me go

(and by that i mean it was some /shitty/ code)

you should return to bed

probably tbh

you should do the tired loafy NOW

be healthy or else 🔪

[zealous tulip]

when looking at what https://youtu.be/yBJJYkC5cdk did I found him using the nightly array_chunks which is actually better since it gives a array and not a slice over them which means you can deconstruct them in a map (and it also works on iterator)

[red pelican]

none of the things to figure out are checkmarked :( https://github.com/rust-lang/rust/issues/74985

why does length 0 even need to be an error?

[zealous tulip]

well

x / 0 chunks....

[red pelican]

actually I was thinking of it in terms of array_windows instead for some reason, where it always advances by one so my bad

asking about it here in the rust community discord #818964227783262209 message

[zealous tulip]

let me just join that real quick

[red pelican]

https://discord.gg/rust-lang-community

I have almost 6k messages there

[zealous tulip]

I know of the discord I just do not like to stay in big ones

[red pelican]

Ok

[livid pier]

Are you all doing yours in one rust source folder? or are you doing cargo new day-01 / day-02 etc for each day?

If the former, are you using a lib or a binary?

[zealous tulip]

I'm doing it as a binary

[livid pier]

I knew my code editor looked weird. like no indication on types of things and i thought I broke something

Its because im in one folder, that has my folders of daily solutions in it

so there is no "workspace" for the rust analyzer to detect stuff from

[zealous tulip]

you can make a workspace if you want one

make a Cargo.toml in the root folder and then do something like

this will make your target folder be in the root too

[tardy barn]

I have an Advent-of-code folder with a default cargo project inside of it, then an alias that makes the day-x project

[frail slate]

members = [ "day-*" ]

[zealous tulip]

damn

My manual effort has gone to waste!

[livid pier]

Saved me though

Unfortunately rust analyzer still unhappy

[zealous tulip]

what's the error?

[zealous tulip]

hm, idk

maybe restart vscode

[livid pier]

Nah, I'll just do something else. Going to just put it all in one folder

I did the google trick and copy pasted and no solution worked for me so far and its like 1 am so really just going to sack it off haha

[livid pier]

(I didn't bin it off) I figured it out

Thank you for the help 🙂

Now everythings a lot nicer

[crude juniper]

Here's my go at today: https://gist.github.com/JELAshford/1c0145edc08096dda841dfe55fc577d7.

I much prefer how you found the priorities @zealous tulip, I was looking for a way to do that (char as u32) conversion but couldn't work it out!

[zealous tulip]

Yours is a bit more intuitive

[livid pier]

That is my solution for part 1, part 2 and the helper function I used

[marble stream]

im doing day 7 of 2015

i think my approach will work but holy moley i am bad at coding

[tardy barn]

Mooood

[crude juniper]

Day 4! https://gist.github.com/JELAshford/7298f9f2e0d7237a9a8c9fc0a3f7e527

Converting a bool directly to int for a oneliner function feels bad...somehow? But altogether the code run in ~7micoseconds without prints, so I'm happy 😛

[tardy barn]

Here's my day 4 utter and absolute spaggheti

It's such trash

BUT

It works

[frail slate]

why do you || collect || so much 😭

[zealous tulip]

https://github.com/Coca162/advent-of-code-2022/blob/main/day-4/src/main.rs I'm going to be honest I did the second challenge through advanced guessing

[frail slate]

shit i wanted to use || split_once|| but i didn't find that function 😭

[zealous tulip]

oooof

it's very nice apart from the ||tuple deconstruction hell, I was going to use ranges but then I realised that wasn't needed for challenge two (or well I couldn't find a helper function which does contains on two ranges)||

[frail slate]

|| i made one of the ranges an iterator, and found whether each value iterated is contained in the other range ||

[zealous tulip]

||yeah that was like what was happening last day so I wanted to not do it again||

[zealous tulip]

||Annndd I realise that my filter for the second challenge was overly complicated, it just needs to be first_max >= second_min && second_max >= first_min||

[tardy barn]

• Problem: I need Vector
• Solution: Collect

[tardy barn]

(Suggestions to de-spaghetti-fy are very welcome)

[zealous tulip]

||you can use split_once instead of split||

[tardy barn]

Ohh
i forgot what it was called but I knew it existed
Thanks

[modern delta]

my solution for day 4, i tried and use more closures :D

https://github.com/eriizu/aoc_2022/blob/main/src/days/d04.rs

[zealous tulip]

Isn't from_str_radix(input, 10) the equivalent to from_str(input) 🤔

and thusly from_str(input) is the equivalent to input.parse()

[modern delta]

maybe

i'm still new to rust

[zealous tulip]

it seems like they do use from_str_radix(src, 10) to implement FromStr using a macro for all integer types

[modern delta]

I'm definitely using from_str_radix directly to save compile time /s

[zealous tulip]

mhuahaha

[red pelican]

day 4 solution ||https://pastebin.com/CU1U2KaH||

runs in about 100 microseconds on my computer, without file reading and printing

I should have used ||RangeInclusive, not Range|| tbh

[zealous tulip]

I mean you don't need to use a range at all

[red pelican]

;compile rs fn main() { dbg!(std::mem::size_of::<std::ops::RangeInclusive<u32>>()); }

[undone creek]

Compiler Output

[/app/example.rs:2] std::mem::size_of::<std::ops::RangeInclusive<u32>>() = 12

TheRoboMan#5720 | 57ms | rust | rustc 1.65.0 | godbolt.org

[zealous tulip]

huh

[tardy barn]

Wha

[red pelican]

its a shame, one of the worst rust std design decisions that made it to stable

ranges directly implement the Iterator trait. that makes them bad at being just ranges, they should be only IntoIterator

[zealous tulip]

;compile

fn main() {
    dbg!(std::mem::size_of::<std::ops::Range<u32>>());
}

[undone creek]

Compiler Output

[/app/example.rs:2] std::mem::size_of::<std::ops::Range<u32>>() = 8

Coca162#6765 | 33ms | rust | rustc 1.65.0 | godbolt.org

[zealous tulip]

an extra 4 🤔

why

[zealous tulip]

yeah :/

[red pelican]

pub struct RangeInclusive<Idx> {
    // Note that the fields here are not public to allow changing the
    // representation in the future; in particular, while we could plausibly
    // expose start/end, modifying them without changing (future/current)
    // private fields may lead to incorrect behavior, so we don't want to
    // support that mode.
    pub(crate) start: Idx,
    pub(crate) end: Idx,

    // This field is:
    //  - `false` upon construction
    //  - `false` when iteration has yielded an element and the iterator is not exhausted
    //  - `true` when iteration has been used to exhaust the iterator
    //
    // This is required to support PartialEq and Hash without a PartialOrd bound or specialization.
    pub(crate) exhausted: bool,
}```

[zealous tulip]

bruh

[tardy barn]

Lmao

[red pelican]

specialization would be great, even a limited version of it

specialization is basically the ability to have a blanket implementation and another implementation overlap

[zealous tulip]

it seems your pc 2/3rd faster then mine

[red pelican]

afaik my cpu is pretty good but gpu is not

[zealous tulip]

who even needs gpu

what am I going to do with a fancy gpu! run factorio at a slightly higher quality???

[frail slate]

then you would need to duplicate code

for Range it makes no sense

for RangeInclusive i dont understand the extra field

[zealous tulip]

I made a alternative way to parse the input which is very funky

[red pelican]

I think ive heard its because of for n in 0..=u32::MAX

how do you tell when to stop iterating

[frail slate]

hmm

fair enough then

[frail slate]

that's what i did lol

[red pelican]

what I did||rs fn parse_line(line: &str) -> Option<(Range<u32>, Range<u32>)> { let mut iter = line.split(['-', ',']).flat_map(|s| s.parse().ok()); let ranges = ( iter.next()?..iter.next()? + 1, iter.next()?..iter.next()? + 1, ); if iter.next().is_some() { return None; } Some(ranges) }||

[zealous tulip]

idk I like working with the tuples

repeatably calling .next feels a bit jank

[frail slate]

its only twice

[zealous tulip]

true, ||but you are flattening everything and then I assume chunking it back into pairs||

[frail slate]

yeah but || i used Itertools which chunks back into an iterator so i had to do .next() anyway lmao ||

[zealous tulip]

ahhh

dammit I need array_chunks in my life

[zealous tulip]

https://crates.io/crates/nom might use this tomorrow

[livid pier]

Day 4 - Part 1 and 2 solutions

This is my string to int converter function that I threw together just to look a little nicer

[frail slate]

i always wanted to learn nom and this is the perfect opportunity

[zealous tulip]

nom nom nom

[mental owl]

CUTE

waa i didnt do yesterdays one

[frail slate]

today's puzzle was really fun

https://kaylynn.gay/blog/post/rust_ranges_and_suffering

relevant to the discussion before

[red pelican]

nice, to switch to solving part 2 of day 5, I only had to ||remove a single .rev()||

[zealous tulip]

Tfw I get frustrated with nom and then ask openai to give me a clue

and it gives something more informative then 10 minutes of googling

[zealous tulip]

||To note I had no clue what map_res was since it was a modifier and I only looked at the combinator table||

[dapper abyss]

i know i am late : rubberduck: but i think that a games is for 3 rounds does not matter for the overall score, does it?

day2

[mental owl]

.net standard library has Too Much Shit

week 1 gonna be autosolved by Framework Types for me

maybe i should have started doing this in netstandard

just to spice things up

[mental owl]

got to today's tasks

parsing input looks annoying, everything else not so much

[zealous tulip]

I am currently in nom hell

[frail slate]

nom was so good

some of the functions are weirdly named so I struggled to find them

[zealous tulip]

i like using it A LOT

it's just the learning curve is getting to me

[frail slate]

it's not too bad, it's just HOFs

[zealous tulip]

I messed up ||because I realised I actually needed to make the crate parsing function return a Option<char> so that the vector positions match the actual position||

[frail slate]

|| i just used the space character lol ||

[zealous tulip]

||hm, I think i'll keep it as a option for reasons||

[red pelican]

https://fasterthanli.me/series/advent-of-code-2022

[zealous tulip]

indeed

I have so much aoc stuff to explore after this is done

[red pelican]

isnt reading those a good thing to do just after youre done with the one problem? you could use it to see how you can refactor and improve your code

[zealous tulip]

I was doing that with https://www.youtube.com/playlist?list=PLWtPciJ1UMuBNTifxm5ADY65SkAdwoQiL

so now I got 2 of them to read

maybe at the same time

[tardy barn]

How'd you guys even || get the crates' order from the input||?

I just hard-coded them

Most cringe way to code it possible

[zealous tulip]

Pain

I am so close to finishing parsing

I forget what the actual challenge was

[mental owl]

stack strings are consistent in length

so like, you can dumb-index it

or are you all trying to do it in an inputerror prone way on top of doing it in rust

[tardy barn]

C#
shudders

[mental owl]

my rust compiler got fucked up okay

besides ive never done aoc before

[red pelican]

I did the same. honestly it doesnt feel like that's meant to be part of the programming problem to me

[red pelican]

spend that time making sure your code is right or making a unit test maybe

[zealous tulip]

how dare you suggest a good solution

[red pelican]

ive been making unit tests for 3 days now, to cover at least most of the examples in the problem instructions

https://doc.rust-lang.org/stable/book/ch11-01-writing-tests.html just in case

to make it easier to check it out if someone reading this doesnt know how to write tests in rust

[mental owl]

someone i know is doing this aoc in python oneliners

[zealous tulip]

so that's the issue

fuuuu that was the issue

[mental owl]

ratio me all you want

[zealous tulip]

||instead of a new line I separated with a space||

What a nightmare

I'll publish the code soon

[zealous tulip]

a https://github.com/Coca162/advent-of-code-2022/blob/main/day-5/src/main.rs

[livid pier]

Me too

Well slightly different but same same

[red pelican]

why not vec! macro

[livid pier]

That is what I did

[tardy barn]

Forgot it existed

[mental owl]

what's the mark after which AOC tasks usually get wacky?

i assume 1.5-2 weeks?

[zealous tulip]

Hopefully not in 5 days

I won't even be able to do the one on Saturday

So I'll have to do two on Sunday

[crude juniper]

Just got it done on time! 🙂 https://gist.github.com/JELAshford/807c71520d7a63a8fc37e27baf0bfb20

[zealous tulip]

There is technically no time limit

But it actually unlocks the new one at 5 am ;)

Or well midnight for EST

[crude juniper]

That was nice! Sure it can be done faster, curious to see what you all do https://gist.github.com/JELAshford/a4933fcad39f0501a48fca0a8a0aea4f

[mental owl]

from last year

[mental owl]

does wait time increase with each failed attempt?

[red pelican]

ok im not going to be doing advent of code at least today. It's just draining my energy that I currently need for other things

[frail slate]

Today's is pretty simple in Rust. Both parts took like 10 lines of code

[zealous tulip]

Yes

[livid pier]

Cleanly 10 lines of code? Because boy howdy I am not seeing it in 10 lines.

Hilariously, I find iterating over/parsing strings in Rust so infuriating.

[frail slate]

approximately

there isn't much to parse here lol

[livid pier]

I think I might just be too fond of the hack and slash capabilites of Python.

lol I'm going to give it a genuine go in a bit but right now I am just reading the instructions

[zealous tulip]

I really like how iterators make me think rather then making everything a for loop

[livid pier]

I finished part one. ||and the way I wrote my code for part one, means all I had to do was change one parameter and it worked for part 2 ||

Day 6 Part 1 and 2

[zealous tulip]

https://github.com/Coca162/advent-of-code-2022/blob/main/day-6/src/main.rs tried making mine as funky as possible

||VecDeques are pretty cool tbh, I like the idea of a cyclic vector, also the next_chunk function is amazing for giving me a straight array and not a slice, also hadn't used constant numbers before this but they are cool||

[zealous tulip]

oh god today is really hard

[mental owl]

ohhh this is exactly the thing i like doing i think

[zealous tulip]

I am going to try make my solution more generic because if I try to only solve this problem I would have to rewrite everything

[mental owl]

waaaa

i dont feel like doing this today

lazy ass wasp

[zealous tulip]

It's hurting my brain

[mental owl]

||i made a parser recently so it shouldn't be terribly hard... i just feel lazy overall||

[zealous tulip]

||I've been making parsers since the start of december! /s||

[mental owl]

nevermind i woke up somewhat

result shouldn't include root folder right

this could use some slightly clearer wording,,,

[zealous tulip]

i knowww

I am crying right

now

[mental owl]

just caught up with everything but todays... today's is too complicated and im still ill not gonna happen

[zealous tulip]

I have a bug and it is breaking everythinggggg

[mental owl]

i have a bug...it's making me cough a shit tonne

ba dum tss

[zealous tulip]

I am currently on 6 tabs of indentation

[frail slate]

cease

[mental owl]

part one done

||i went a very naive tree structure way but it works||

[mental owl]

the bit i did took me to 9 tabs, but i deleted it because yk

[frail slate]

i did the parsing and gave up at the tree stuff

[mental owl]

i was three match statement and four if statements deep

[frail slate]

skill issue

[zealous tulip]

I am close to finishing

I think

||currently just trying to flatten my "tree" into all the directories but I don't think it's working||

FUUU

OH IT'S AT MOSTTT

not at LEAST

[mental owl]

ooopsie daisies

[zealous tulip]

it did not cause that much pain

thankfully

thank god

I'M FREEE

https://github.com/Coca162/advent-of-code-2022/blob/main/day-7/src/main.rs my abomination of a solution

[mental owl]

bruh, VSC official C# plugin doesn't have inference or hints for call chains like that

[zealous tulip]

yeah

I was surprised when I first saw it in rust

[mental owl]

fucking what

i'm discovering new things today

[zealous tulip]

tf

[mental owl]

if it means what i think it means that's

INCREDIBLY idiotic

and i hope it does not happen

[zealous tulip]

just change all of the dotnet standard library

it'd be easy!

[mental owl]

🥹

[mental owl]

meh, that's a lint i like. if it's idiomatic to do it otherwise and it might be a future conflict then why not add it to the linter

worse than bullshit naming conventions are enforced bullshit naming conventions

oh wait it's enforced?

i mean

sure sounds like it

[zealous tulip]

idk about that

[mental owl]

yellow line ≠ warning???

yellow line is warning yes

okay so it ain't enforced (unless Werror is a thing in C#?)

it's not as of now, but since "reserved" usually means "no one else is allowed to", i assume there's a proposal somewhere to make it enforced

maybe i assume wrong

all that reads to me is "don't bother being un-idiomatic here, we use this system to name keywords or std or whatever (im not a c# dev)"

[zealous tulip]

"we MIGHT use these in the future"

[mental owl]

i mean, rust reserves a few keywords for unimplemented features but it can never be exhaustive, i prefer c# here in that my code wont have a conflict ten years in the future than rust being like HOPE ITS OKAY

yeah probably

i just had a flash of paranoia i guess

[livid pier]

I'm fuming, all the tests work but my answer is too low

[zealous tulip]

Small tip ||Check that you can handle multiple cd .. correctly||

||Well in general it's best to look at what is being done in your data that isn't being done in the sample||

||also a debugger can be a good idea but you need to learn how to use it a bit beforehand||

[livid pier]

||I am handling multiple cd .. properly ||

[mental owl]

actuallu no i'll move it

[zealous tulip]

Welp shit

There is so much that can be wrongggg

[livid pier]

I know

I am just in hell right now lol

I abandoned Rust after like an hour and a half

I have been in Python for the last 2 hours

and I am just staring at the void

[mental owl]

python moment

[livid pier]

I thought it would be easier since I've used it more but somehow I've gotten myself more twisted and its now 1 am

[zealous tulip]

😭

Drop it it's not worth it

[mental owl]

i wonder if i would be in as much pain as you folk if i wasn't using my comfort lang

[zealous tulip]

It would not say it was painful

Annoying and tedious? Maybe

Also I kind of don't know if I like C# iterators anymore

Though the borrow checker was actually kind of a pain for this one

||Since tree structure does that to a mfer, I tried to store a mutable reference to the parent and it did NOT WORK AT ALL||

Now I hope tomorrow it won't be as difficult

[mental owl]

idk

yielders for life

[livid pier]

I am in hell. I checked everything. I have spent a while testing things. I ran my friends code. I got the same result. The same wrong result.

The test input gives me the correct value

||95437|| for the test on part 1

IDK why i blocked out the answer for the test lmao

Can someone validate that this is the correct input file?

[zealous tulip]

They are made for every user

So no

[livid pier]

Ah my god I am dying lmao

But wait

that infuriates me even more. His code and my code resolve to the same answer

[zealous tulip]

Well that's not ALL your code||, though what's recwalk||

[mental owl]

||recursively iterates over a directory's children||

||all my code: https://github.com/thalber/aoc-2022/blob/main/CS/d7_cs/Program.cs||

[zealous tulip]

||I find it funny how he had to make a part 2 https://youtu.be/lRhZJIy8zhE||

[zealous tulip]

||That yield is pretty awesome ngl||

||we need rust yields 😭||

[mental owl]

||when making fs, i instinctively started overengineering and had to hold myself back like "hold on, full path cd is not in task spec, i probably should not add things that won't be used"||

[amber siren]

[pt] i'm pretty sure rust's approach is actually even more backwards compatible
new strong/reserved keywords that weren't previously strong/reserved can only be introduced in a new edition, which is the only time breaking changes¹ are allowed at all
code from different editions can be used together without anything being different at all (changing the edition of a library doesn't break its users in either direction)
anything can be used as an identifier by putting r# before it

¹rust allows some things that are technically breaking in some circumstances to be minor, stuff like creating new things that are public, implementing a trait, making the signature of a function more general,
with the idea being that you should be able to work around minor changes just by adding extra annotations

[frail slate]

i have completed today's but im still crying at yesterday's

[crude juniper]

Same 🥲

[mental owl]

sin

[zealous tulip]

shit I cannot do part 2

I think my lack of experience with algorithms is starting to show

[mental owl]

haven't gotten to the tasks yet, but ||i think building a visibility cache grid should be good enough here||

[zealous tulip]

||I have no idea what you mean by that||

[frail slate]

My solution is very hacky lol

[mental owl]

||concept: for each row/column, in each direction (u/d/l/r), you can take a tree height and determine where is there a first tree that would shadow it. ie in the example||

30373
25512
65332
33549
35390

||for row 1 you'd remember: height 0-3 is shadowed starting with position 2, height 0-7 are shadowed starting with position 5. I'm going to cache these on one loop and do something with cache on another, whenever i get around to it||

[zealous tulip]

what the fukkkkk

[mental owl]

?

[zealous tulip]

It's just not something It'd have thought about

OH that gave me a interesting idea

[mental owl]

ughb my brain is not working right now

no aoc for a few hours

[mental owl]

✨

[zealous tulip]

Beyond COMPREHENSION!

[mental owl]

i will be slightly surprised if this clump of code works first try

huh, it almost worked

20 instead of 21

(probably something gone horribly wrong somewhere and it's an accident)

[zealous tulip]

20 + 1 boom fixed

I mean if that solution works for different inputs.... then is it a wrong solution?

[mental owl]

guh

i lost motivation again

[zealous tulip]

I am getting really concerned because I will not be able to do the aoc for the next 2 days so I will have a backlog that i probably won't be able to catch up on

[visual spire]

shouldda pinned this before, it's now pinned

[livid pier]

I am toasted TBH

I again am stuck. I thought I had all edge cases worked out. I have tests working perfectly and then I run the full input and it broke

[mental owl]

love it when that happens

[livid pier]

Honestly, i don't!

[frank cairn]

i think that was sarcasm

[livid pier]

I'm going to take a shower and give it a swing one last time but it's almost 1 am so I think this will be a RIP for me

[mental owl]

last time it did that i just switched to a solution with half as many iterators and twice as many LoC

[livid pier]

I switched programming languages

First time I ran the code it worked

I went from Python, to Rust.

[livid pier]

Progress so far

The pivot from trying python and struggling last night and having my rust solution work first try has got me back following the light of Ferris

[zealous tulip]

https://twitter.com/njvsvh/status/1599133081507856384

igor vaiman (@njvsvh)

glad to see I'm not the only one:

(blue) dec 1, AoC started

(red) dec 3, the first challenge requiring hash sets

#AdventOfCode

Likes

387

Twitter · Dec 3, 2022, 8:06 PM

[frail slate]

lol

[livid pier]

Day 10 Part 1:

||Honestly was prepping for new commands in part 2 or additional signal checks or whatever and NOPE||

[livid pier]

Day 10 part 2 - Only thing slightly cut off is the print capability for the answer

[crude juniper]

Hey all! I'm catching up on AoC today/tomorrow and I'm hitting a Rust-related speed-bump. I'd like to visualise the current state of my knot-moving for Day 9 so I'm creating a Vec<Vec<&str>> to insert characters into at the 2d coordinates I know have a knot in. As part of the visualisation, I want to number the positions on the knots as per the problem spec. As I iterate over my Vec<_> of position object, I use a "usize" index to work out which number to put in that place. All that said, the issue I have is I can't find a way to convert that "usize" index into a "char" without going through "String", and that leads me to this error:

error[E0716]: temporary value dropped while borrowed
   --> src/main.rs:101:38
    |
101 |             let mut display_char = &*((self.sections.len() - 1) - ind).to_string().as_str();
    |                                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^         - temporary value is freed at the end of this statement
    |                                      |
    |                                      creates a temporary which is freed while still in use
...
105 |             output[(position.y - min_y) as usize][(position.x - min_x) as usize] = display_char
    |                                                                                    ------------ borrow later used here
    |
    = note: consider using a `let` binding to create a longer lived value

To the best of my understanding, it's the fact that I'm making a String for the "char" to reference which then gets freed afterwards that's the problem. But I can't find another way to go "usize"->"char"! Any advice much appreciated; code context for error in attached picture.

^and I just noticed the missing semicolon on 104 😛

[frail slate]

https://doc.rust-lang.org/std/primitive.char.html#method.from_u32

is this what you want?

(i haven't done the last few days)

[crude juniper]

Yep, that's the one!

Looks like I was getting "char" and "&str" a but muddled up too

Thanks muchly!

Ah but also no, it seems - converting usize->u32->char like this doesn't quite do what I expected.

[livid pier]

I dont know what has happened but I am frozen. My computer is stuck on this:

cargo run -p day-11
Blocking waiting for file lock on package cache

I have ran cargo clean, I have removed cargo cache files, I have restarted the computer. I have no idea what could be causing it

I have even closed vscode and everything, tried to run from iterm2, default terminal, etc

I even killed and restarted the cargo process

[livid pier]

Took me forever but got it and finished Part 2.

||Did lowest common multiple to not have to use arbitrarily large numbers.||

[livid pier]

I had to tap out on Day 12.

I am officially going to be a day behind from now sadly

Going to try and finish it tomorrow

[nova mango]

I haven't done today's or yesterday's yet because I've been sick

[modern delta]

the way covid hit me

[frail slate]

😭

i got lazy/unmotivated

[mental owl]

same

laundmo destroyed me with minecraft

i can no longer do something that isn't minecraft

[zealous tulip]

the day of reckoning (day 8), though it was actually because I had to go on a 2 day long travelling session

[mental owl]

well

I'm not obsessing over Minecraft anymore but i also don't have the motivation to start aoc again

fuck

[modern delta]

Just did day 10 and I had such a hard time understanding the behaviour for step 2 😭

[young knoll]

ive tried getting into aoc multiple times but its like

my brain doesnt have the processing power to digest the walls of text into the actual problem at hand

[red pelican]

yeah its a little annoying. maybe just dont even try to analyze any logic until youve read it to the end

[young knoll]

no its like

my brain cannot process it it just processes it as background noise its like "oh the elves are staring at this pile of lava because the raindeers are stuck and-" and im like "what does any of this have to do with programming"

"just tell me what i need to do"

[mental owl]

usually the task starts after the words "(your puzzle input)"

[mental owl]

back on day 8

my idea of how to do it is fine, but i'm struggling to come up with an implementation that wouldn't be a nightmare of flipping indices to write

[livid urchin]

What is Advent of Code?

[zealous tulip]

daily challenges with increasing difficulty during October and until christmas

[frail slate]

October? 👀

[zealous tulip]

very.

[red pelican]

Very daily very challenges with very increasing very difficulty during very october and until very christmas

Its only 1st of december to 25th of december (at least this year, idk if it has been different other years, I really doubt it considering its called advent of code)