advent-of-code

[humble copper]

close to burning out, really hope the next day is easy

[north silo]

I didn't actually intend to stay up for this, but here I am 👀

[humble copper]

(its still today for me bc it unlocks 9 pm)

[woven sable]

I think Eric should be aware though that tomorrow is Christmas Eve and so hopefully it's easy

[humble copper]

ive heard he makes it easy for christmas day

i hope that also applies to christmas eve

watching the aoc conference that eric did in 2019

the crowd is just dead silent the whole time

(not literally, but still)

[woven sable]

~30 seconds left

[woven oriole]

.aoc c

[hidden copper]

Day 24 starts <t:1640322000:R>.

[woven sable]

10 seconds left!

[hidden copper]

<@&518565788744024082> Good morning! Day 24 is ready to be attempted. View it online now at https://adventofcode.com/2021/day/24. Good luck!

[woven oriole]

oops

[lost dock]

NICE

[crystal shard]

do you live in this channel;

[woven sable]

OH MY

today is amazing

I love it

[foggy zealot]

sixth

[crystal shard]

seems fun

[languid stratus]

I feel like once again, my lack of understanding of the question in the first place might be my downfall here

I should probably turn my screenreader on but I'm not sure it'll like all of the numbers etc.

[crystal shard]

wow 8 mins

[humble copper]

ERIC

THANK YOU

[devout tusk]

i don't even understand how I will even approach today's question 💀

[woven oriole]

seems like no one understood the ques lol

[devout tusk]

p2 is defo going to have jumps

[woven oriole]

bruh where is the number

it is only initializing variables

and doing operation

[devout tusk]

inp is the number

[woven oriole]

its value is 0 right

hm

[woven sable]

alright I think my sol is odne maybe

it's just real slow

:P

[devout tusk]

no, value is non zero

[humble copper]

||pretty sure it wont, the input doesnt change between parts||

[woven oriole]

The ALU is a four-dimensional processing unit: it has integer variables w, x, y, and z. These variables all start with the value 0

[woven sable]

someone finished both parts already, damn

[devout tusk]

just like yesterdays...they may tell to add some small shit to input

[woven sable]

anyone else use ||match|| & ||case|| forthis?

[humble copper]

||nopeeee 3.7.3 all the way ||

[sonic silo]

||im in 3.10 but i dont wanna use match case||

[humble copper]

actually this is a lie ||i use 3.8 now bc it's 64 bit ||

extra tempted to ||just use a dict to store the values of everything||

perhaps ||the forbidden eval()|| will make a reappearance today?

[languid stratus]

I have a working code but I have done some calculations and it will take roughly 3.17 years for it to run fully

[humble copper]

well this is.... ||a bit large....||

[humble copper]

for this, can we assume that ||all values below the maximum are valid and all above are invalid|| or ||are the valid values scattered throughout the possible values||?

[crystal shard]

you did it by hand?!?

[humble copper]

||9^14, 9 possible values per character, 14 characters required||

[prisma sundial]

🙂

[humble copper]

nice

||so part 2 is easy?||

[prisma sundial]

part 2 is min instead of max

so yes

[pearl moon]

So the crux of this is to look at the || instructions that follow each input and figure out the valid input range for that digit|| right?

[humble copper]

lmao

what

haha thanks eric

you're a lifesaver

[humble copper]

||are the valid values scattered or are they constant up to a certain point||

[prisma sundial]

i have no clue

[pearl moon]

Could a runtime for the solution be || 14 * 9 aka O(n) in the number of digits of the model number ||

[humble copper]

keep in mind ||x y and z stay the same after inputs||

[pearl moon]

Sure, so you have to || manage that carryover as well ||

[humble copper]

so ||for setting x to x + input you'd have to also take into account what x was before an input||

i was thinking ||find z as an expression of inputs a-n||

then ||pray that eric made the valid numbers constant and not scattered ||

i'll do this tomorrow, but im really glad it's this easy

[earnest forge]

I literally just wasted a minute by typoing 14 to 24 -_-

[earnest forge]

Yes, you should really be ||decompiling your input||

[lapis beacon]

.aoc unsubscribe

[hidden copper]

Okay! You have been unsubscribed from notifications about new Advent of Code tasks.

[mossy stream]

finally a fun and quick puzzle

part 2 better be nice

[daring monolith]

.aoc unsubscribe

[hidden copper]

Okay! You have been unsubscribed from notifications about new Advent of Code tasks.

[dense herald]

For day 20, part 1, I get the same output for the example after the two image enhancements, but an incorrect result on the real input. Some of my assumptions are that the image expands by 1 pixel on every side with each enhancement and that every new pixel value is determined before any existing pixels are altered. Is this right?

[restive imp]

I think your assumption is somewhat right. This is what helped me: #advent-of-code message

[proud cosmos]

time to figure out something smart

[ruby epoch]

For AoC day 5, I'm handling those coordinates by checking if either side are equal and if so, taking the other side and using them as a range to append to a Counter, basically doing for i in range(y.1, y.2+1) when x is equal

However, when I hit the line 2,2 -> 2,1 from the test data, I'm basically doing for i in range(2, 1+1), which is clearly an issue, so I'm trying to work out a better way to handle this. Has anyone got any ideas? (The screenshot shows the portion of code I'm looking at)

[woven solar]

You can just swap the two if they are in the wrong order.

e.g. a,b = min(a,b), max(a,b)

[ruby epoch]

That sounds like it'll work. I never think of those

[proud cosmos]

Turns out to be more work than I thought to get a smartish solution

[muted aurora]

Haven't looked at today's thing yet, how is it in terms of difficulty?

Compared to the last few

[frigid mural]

naive solution is trivial, its just an optimisation problem

[muted aurora]

So like most other days then

Cool, thanks

[proud cosmos]

Hmm

Still not sure how to optimize it

seemed easy but maybe it's not that easy?

Maybe i'm jst dumb and doing useless stuff

Lol

[mossy stream]

... there is no example for today, is there

[proud cosmos]

There are a few

just no working value

Hmm

Honestly copnsidering just doing it on paper

seems easier than trying to find an algorithm at this point

[candid berry]

Hard. 😄

[mossy stream]

I think I underestimated the time needed to bruteforce

I don't want to do constraint solving

[frigid mural]

||imo the main idea for optimising is the fact that whatever an input digit is, its not equal to a variable if its 10+, which leads to a lot of cases where a digit is free to be whatever||

[proud cosmos]

I feel like I saw that but I am still not sure how to actually use that information to do anything

I guess one more thing to reduce

[frigid mural]

and thats why im going line by line with pen and paper 🎉

[proud cosmos]

yea almost feel like that's easier

than doing it with an algorithm

just get a z = at the end or smth

[frigid mural]

||trust me there's a TON of useless operations||

[proud cosmos]

lol

Yeah

||I looked at the start and the first few are already useless||

Lol

[frigid mural]

||im 45 lines in and there has been a net of 1 useful operation||

||all the other 'useful' ops are literally part of something that resolves to 1 for whatever input digit so no need to even calculate em each time||

[proud cosmos]

I wanted to make somethign that goes through and figures this out but it didn't work out so well

maybe its no-code time

[frigid mural]

Oh wait big brain idea

[proud cosmos]

nice

lets hope it works out ebtter than my ideas

they haven't so far

[frigid mural]

||We could instead go through sections at a time, say between y=0 resets, and run through the possible digit inputs, link them to the output variables. If the output dict is constant then we know the value of the digit doesn't matter so we lock that in at 9 and move on||

[old sentinel]

hey

so

[frigid mural]

||Would still have to manually go through the actually meaningful digits if we want more optimisation but still||

|| oh duh the splits we wanna use are just the input lines, so that even if the w variable changes then we ignore it anyway as it's about to be overwritten||

[proud cosmos]

I kinda tried something ismilar to this but I think it doesn't quite work

Maybe I was just doing it dumb though

I think I tried something different with those that makes less sense

tbh

[frigid mural]

||I'm sure there are still exceptions and you'd have to take a look at them, in fact there necessarily must be orher wise 999999...999 would be valid||

But cute down a lot of the work

[proud cosmos]

It's possible it may cut down the work yea

[frigid mural]

||Even a single deduction divides the search space by 10 so whilst we may not get them all we still have a good shot, and can perhaps manually finish it or run exhaustive search||

[proud cosmos]

I've started loving asserts tho

more asserts more better

I've also noticed that || each of the sectors has barely any difference, 1-3 commands wonder if that is useful to know||

[uncut wave]

#discord-bots

[frigid mural]

                    case _:
                        assert False, 'AAAAAA'
``` in my match-case tree, just for you

[proud cosmos]

Nice!

I'm back in action lets try this again

[primal vapor]

are your end z very big too ?

I have things like 196685829

[proud cosmos]

Yea

Still trying to figure out how to limit anything

except no-code

which sounds easier at this point

Though I should be able to do what I do with no code in code as well right

[primal vapor]

yes

it's frustrating

[proud cosmos]

Seems hard somehow idk

maybe I just need to start blank

[frigid mural]

||if you cut the code into blocks you can abuse the repetition of patterns in there to reduce the computation required for each step||

[primal vapor]

did you successfully brute force it after that change ?

[frigid mural]

not yet but i can see how itll work, as there's ||a lot of cancellation, for example only the prev. value of z is relevant in each block||

[primal vapor]

yes I saw that

[proud cosmos]

Yea

I'm thinking if you can solve each block based on requirements

That should technically work sounds pretty hard tho

[frigid mural]

||could go backwards block by block too ig once you know the relations||

[proud cosmos]

||Like if you do the math you can find out that the last block either requires z to be 0 or z%26 to be equal to w, you could solve the other blocks trying to get either of that||

[frigid mural]

exactly

[proud cosmos]

Just not sure how you'd get that with code

Lol

Does not sound trivial

[primal vapor]

yup

[frigid mural]

||looks like it probably wont ever be 0, so you can probably get away with using mod26 throughout||

[proud cosmos]

Persumably

[primal vapor]

I just got an idea

reading a reddit post

||"The number of possible states the ALU can reach at a given line is significantly smaller than the possible number of input read "||

so maybe

[proud cosmos]

Idk ||I tried going that route and the first sector literally just sets z to w+7||

||Which would make the states = to the inputs||

Maybe it gets betters

[primal vapor]

||we can build a dictionary with key (z,w,current_digit_index) and value true if final z is 0 and false if it's not||

so that we don't calculate 2 times the same thing

[proud cosmos]

basically what I wanted to do but it seems to still have a ton of possibilities so yeah idk

[gusty basalt]

hey guys my input code is correct but vs code is not lettin' me enter the input how to resolve this error

a = input("Enter your age\n")
if(a>=18):
    print("Yes,your age is greater than 18")
else:
    print("No,your age is less than 18")

someone pls help me out

[proud cosmos]

rofl if this pattern repeats the entire time I am pretty sure i know what to do to calculate it faster

no-code op ig

[north silo]

Yeah I have no idea what I'm doing today

[proud cosmos]

Stare at it for a long time and then continue being confused

smth like that

Feels not that hard though

[woven oriole]

if you ever get the idea what you are doing care to explain

https://cdn.discordapp.com/emojis/739614945045643447.png?size=44

[proud cosmos]

Hardcode never hurt anybody

[livid gull]

onyl person it hurts is people trying to steal your code 😛

[proud cosmos]

And myself a bit

but yea

seems easiest way to get that stuff

[sleek cave]

all of the puzzles starting from d18 are just on another plane of existence lmao (except for d20)

[proud cosmos]

Scientist only know 3 planes of existences so far

normal plane

banana smoothie plane

aoc day18+ plane

We are still trying to get into the smoothie plane

[maiden brook]

I thought day 19 was supposed to be the peak

[proud cosmos]

Apparently not

Though it honestly does not feel that hard today

you can technically solve it trivially not sure if it'd finish today tho

[livid gull]

depends what youre better at, certain problems can be harder for certain people

[sleek cave]

same. then came d21p2, then d22, then d23...

[proud cosmos]

Just rent a few servers..

:)

[sleek cave]

lmao, you can solve d14 trivially, simply wait for a few thousand years, no big deal

[proud cosmos]

Unless you have a lot of RAM/Space tho..

also you can't really parallize that well

today would be trivial

Lol

[woven oriole]

tomorrow should be ok

[proud cosmos]

Could easily split it into 100 servers and search all of them

[primal vapor]

ok aurora

remember the dic thing ?

[proud cosmos]

Hm?

[primal vapor]

forget it

[proud cosmos]

Yea?

Lol

[sleek cave]

pure bruh moment

[proud cosmos]

rip

I got something else now

a magic formula

|| for any step: z=z//v1*(25*(z%26+v2 != w)+1)+(w+v3)*(z%26+v2 != w)||

I believe.

[primal vapor]

itering throught all possibilities is too long even if you just loop without doing anything

[proud cosmos]

There are quite a few

||9**14||

[primal vapor]

yup

so even if you have a magic formula

for each step

[proud cosmos]

Hmm

[primal vapor]

it'll be too long

[proud cosmos]

It must have to do withthat though

and reducing possibilities

[primal vapor]

I think we're not supposed to code x)

[proud cosmos]

You can definitely do it with code

but yeah it kinda feels like it'd be easier without at this point

[wooden jewel]

I am unironically just gonna feed this to ||cplex||

me 👇🏻

[proud cosmos]

Lol

Nice

haha

Beep boop

I'm the computer now

PRob gonna fail horrilby after like 2 or 3 though

[candid sigil]

So day 24 can't reasonably be brute-forced? Scanning the input, I couldn't see any obvious optimizations beyond ||ignoring instructions to divide or multiply by 1||.

[proud cosmos]

We managed to get it down to like a single formula for the value we care about + a few random digits

but even then taht's too much

Even calculating one formula a few quadrillion times would take forever

[humble copper]

by extension, ||if any operation multiplies by zero, just immediately set it to zero||

[proud cosmos]

Yeah there are a few things you can do to reduce it

[humble copper]

or if any ||adds by zero, skip||

[proud cosmos]

ah yes it'll be done soon

[candid sigil]

My ||poorly implemented Dikjstra|| for day 15 took 10 or 15 minutes, which was the longest one for me so far.

[wooden jewel]

I left my day23 part 2 run over night (||A* with quite a few optimisations||)

Guess what

no result

[proud cosmos]

f

[wooden jewel]

Took 32 gigs of ram in the morning

I was ready to punch a mf

[proud cosmos]

Huh

My p2 takes like almost no time you prob had some mistake in it

[wooden jewel]

🤷‍♀️

[candid sigil]

I heard on the subreddit that ||some inputs are a lot easier to solve than others, to the point that one poster had one that was easy to hand-solve||.

[wooden jewel]

@proud cosmos could I try your input?

[proud cosmos]

sure

mine must be incredibly easy

[wooden jewel]

Just wondering

[proud cosmos]

||#############
#...........#
###A#C#B#D###
#B#A#D#C#
#########||

rip formating

But yeah my part2 is incredibly fast with that input

It's like 10x fater than part1

[north silo]

Oh ||I took a closer look at my input and the sequence of operations is periodic, so I think I can take a crack at this||.

Also ||some of the constants in the input seem to be hinting at something...||

[proud cosmos]

HUh?

[woven sable]

You probably have hanging code somewhere

[wooden jewel]

wdym?

It keeps finding new better results, just never actually finishes

[woven sable]

Oh

[proud cosmos]

It should finish relatively quickly tho it prob goes through some invalid states or smth?

Did you see if the example works with the numbers

Maybe assert every step of the example

More asserts more better

Well my hand method worked for about 1/4 of the input

now it;s just too big

:)

[frigid mural]

||the 26 is sadly a red herring if thats what you're referring to:(||

[proud cosmos]

I mena it does kinda have to do smth with it

but not really either

[north silo]

Not looking at this spoiler just yet. I think I'm on the right track 🤔

[frigid mural]

||just if the constant 26 reminds you of the alphabet, its arbitrary and sadly you dont get any nice word outputs||

[bright haven]

that should be spoilered

[proud cosmos]

hmm

I feel quite lost now

maybe I should just let a brute force attempt run anyway

Lol.

[frigid mural]

||the final solution is gonna be brute force at least for me, but you can massively reduce the search space||

||when you reduce the search to 7 digits only you can still do it at a decent speed especially when you know how the blocks work and make that process faster||

[proud cosmos]

Okay I efel like I kinda get what to do now

but I still don't really get how to do it

[humble copper]

i dont

i tried creating a magic equation for z

that i could ||run through eval||

turns out that its too long

and causes a stack overflow

[proud cosmos]

rip

[humble copper]

so i dont know what to do at this point

[proud cosmos]

go ||through the stuff and find a pattern||

[candid sigil]

It does seem ||the sequence of instructions is the same for each block, with only the values differing. So you could maybe write a function with the instructions and partialize it with the values for each block||.

[humble copper]

yep i see it

looking at the input

[proud cosmos]

Yeah and then need to continue somehow from there

^me

anyway

[woven sable]

Or, ||you could use backtracking if you wanted to.||

[proud cosmos]

I feel like I know what my very last number should be

rofl

so that's cool (?)

I think anyway

Like i'm quite certain that it must be that number or else it wouldn't work

I could backtrack no-code from here ig

[humble copper]

so for me ||the values on lines 5, 6 and 16 for each segment of inputs differs||

[proud cosmos]

for me ||[5, 6, 16]||

[humble copper]

wait yeah thats it for me

[mossy stream]

you should have three lines that changes

[proud cosmos]

Okay I can pretty much guarantee that || z <= 5109711748 for all steps||

I can prob press that upper boudn down

might be useful

[humble copper]

how do you get that

[proud cosmos]

||I made a magic formula and checkd what the "impossible best case" would be roughly||

should be way too high though and not acutally doable

Weird wonder why

I mean that's normal though

[ruby epoch]

Any chance I could have some help with part 2 of day 5? I'm finding that this is passing the tests, but when I run it with my actual input, it's lagging out/hanging. I think the issue is with lines 31-36 here (or the whole of the while loop at line 28) but I'm not sure how to go about changing/making it work properly. Any suggestions/advice would be greatly appreciated. https://paste.pythondiscord.com/ulofonanac.apache

Really? I've had whole discussions about previous days here before

[final silo]

Im gonna guess that people are busy those two days usually.

[ruby epoch]

Also where do I go if I don't want spoilers for today?

that... defeats the purpose of a spoiler channel

[proud cosmos]

I feel like it should be fine here, maybe spoiler the link

Though I am not sure how much help you will get most people sit in spoilers when they are done

[woven sable]

Tbf yesterday was quite hard

[ruby epoch]

I'm asking for help on something from 3 weeks ago without seeing the stuff that's happened sinc ethen?

[proud cosmos]

this time surely it's easy

right?

[maiden brook]

we have to complete day 25

otherwise christmas won't be happening

one person* needs to complete day 25?

[proud cosmos]

yes

[maiden brook]

[candid sigil]

I believe day 25 is traditionally easy, because not many people want a hard puzzle on Christmas day, but you never know. Last year it was very easy.

[analog shadow]

what's advent of code

[woven sable]

Advent of Code is an event every year where you get a challenge a day from the 1st of December to the 25th

also you have the exact same pfp as @hidden musk lmao

[hidden musk]

wonder where he got it from

I added the hat, to my screenshot

[woven sable]

Ah

Probably from you then

[hidden musk]

although the character is not mine

[analog shadow]

i was scrolling to discord and i saw ur pfp so i copied it, my bad for not asking permission, is it fine tho if i use it

[hidden musk]

¯_(ツ)_/¯, kinda weird though

[analog shadow]

lol

also what character is it

@hidden musk

[hidden musk]

hayan from unholy blood

[frigid mural]

was a great read, I went on hiatus for a while so im not all caught up

[placid lake]

i have not had time to do the past two days

sad

[woven sable]

Ngl yesterday was pretty hard

A lot of people did it by hand

[placid lake]

yeah, i looked over it

seems.. interesting

kinda reminds me of the towers of hanoi problem

[woven sable]

Yeah exactly, someone visualized it and it looked a lot like the Tower of Hanoi problem

[placid lake]

anywayy, happy holidays to all of you

i’m off for now

[proud cosmos]

Happy hollidays!

[shadow stag]

Only 12h 30m left until the last set of problems are released

[maiden brook]

12h 30m left until we find out what happened to the keys

[proud cosmos]

nvm you've left them at home unlucky

[candid berry]

On each iteration of while tuple(marker) != end_point: print out tuple(marker) and end_point - you'll see what's going wrong

[humble copper]

out of curiosity ||i tried to find the entire equation for z given a whole number. i cant even read the whole thing||

||the whole file is just text with the equation||

[livid gull]

|| you can simplyfy that greatly ||

[humble copper]

yeah i did

but i just wanted to see ||the whole unsimplified equation in all its glory||

[ruby epoch]

I've got to a point where I'm getting a final answer that's wrong. I think it's something to do with the way I'm adding to my ||counter||, but IDK?

[frigid mural]

||throw it into desmos :)||

ez solve

[humble copper]

||its a 14 variable equation||

||maybe symbolab or wra||

[frigid mural]

||only one variable if you use 14 digit x values on graph||

[humble copper]

||wait can desmos do substrings||

[frigid mural]

||no need, just use modulo to separate the digits||

[humble copper]

ok theres just one problem with that

||i cant copy the whole equation because its so big||

[frigid mural]

||the other problem being that theres no way desmos will ever be able to rnder that||

[sleek cave]

taking a break from aoc today 😄 gonna finally start learning numpy

i'm already in love

[humble copper]

i regret never touching this for the entire calendar

[livid gull]

used it for only the cubes i think

[candid sigil]

R is my first love, but NumPy is fast and powerful.

[sleek cave]

the more i read from the documentation, the further i fall in love with numpy

and i'm not even done reading the introduction yet lmfao 🤣

[sleek cave]

same

i'm going to re-do some early aoc puzzles using numpy when i learn it

[humble copper]

wait

question

we've been in a submarine for almost 25 days

in terms of food

...what happened to all those fish we met along the way?

||joke dont get mad at me ik its a stupid question||

[sleek cave]

no grand mission is without sacrifice. collateral damage was inevitable; santa's cause is considerably more significant then a couple trillion lives of unimportant lanternfish. we did spare the snailfish, though, for fear of retaliation... they aren't the best species to mess with.

[proud cosmos]

They multiply too fast anyway

[robust heart]

Merry Christmas everyone!

Still at 28 stars haha

I wanted to give a shot to the other days but I’ll think I’ll take a break for now

A short one though

[robust heart]

I tried to use both when I could

[humble copper]

woah i just did 2016d6 in 10 minutes

thats insane for me

[sleek cave]

ggs!

[robust heart]

Gg

[sleek cave]

usually d4 - 9 takes me 2 - 3 hours lmao

then d12+ takes 3+

[humble copper]

i was watching the advent of code conference and eric said that a college used 2016d6 as the midterm for a computing course

[sleek cave]

then d18+ is hell

[sleek cave]

wow...

[humble copper]

so i decided to do it for no reason

and finished in 10 minutes'

less actually

[robust heart]

I stop at day 10 usually

[humble copper]

i think it was like 5

i wasnt timing

[robust heart]

Also I noticed the personal ranking lol

Got no points because I took 18 hours to several days lol

This is what happens when I’m distracted

[muted aurora]

lol yeah i just tried it, just one-lined it in the REPL

[sleek cave]

2016d6p1 is just 2021d3p1 but easier lmao

almost no difference

[sonic silo]

youd be surprised by how easy college finals can be when it comes to puzzle-likes...

[humble copper]

eric said the final was 2018 days 1-5

so

im trying that next

[sinful pawn]

wait until you hear about Arraymancer

[woven oriole]

Merry chwistmas

🤶 🛒 🛤️ 🎁 ➡️ 🎄🏠🛌

[cursive temple]

santa uses minecart?

[woven oriole]

yes, no sledge in minecraft

[humble copper]

watch eric make day 25 the hardest day on the whole calendar because he thinks it would be funny

[woven oriole]

lmao i too think he has that power

he can ruin the day

😎

[maiden brook]

yes if the keys are lost then christmas is ruined

[humble copper]

given this list of 500 coordinates, find the real cities these coordinates map to, and give the lowest possible cost to travel city to city, assuming each day a different city is visited and, and each night you must stay at a hotel in the city you just entered.

[maiden brook]

that sounds like christmas

[candid berry]

part 2: Assuming you have COVID at the start of your trip, how many people will have it at the end, given the populations of each of those cities and the sizes of their airports, and assuming the Omicron variant of SARS-CoV-2 infects one out of every thousand people exposed to it each day.

[hidden musk]

part 3: find the bomb I've hidden in your house, what is it's serial number

[humble copper]

part 3: given the publicly available covid data, determine the number of variants that exist, and multiply that by the highest r-naught among all variants

[woven solar]

reverse-engineer the bomb's cryptographic handshake and determine the disarm code. You have 3 tries before the bomb explodes as an anti-tamper mechanism.

[humble copper]

star reset

is that even possible

[steel bear]

Happy Christmas at least in asia 🙂

what are you guys doing after AOC ? Leetcode?

AOC was really fun though learnt graph traversal, dijkstra and lots of code golfing XD

[woven oriole]

last time we all will summon

☮️

[hidden copper]

<@&518565788744024082> Good morning! Day 25 is ready to be attempted. View it online now at https://adventofcode.com/2021/day/25. Good luck!

[humble copper]

merry christmas east coast!
¡feliz navidad!

[potent pumice]

Due to strong water currents in the area, sea cucumbers that move off the right edge of the map appear on the left edge, and sea cucumbers that move off the bottom edge of the map appear on the top edge.

[hidden musk]

input the answer into the keypad to disarm the bomb, you have 30 minutes

[pine tiger]

way too close

[peak dock]

lol are d25 part 2s always like this

[woven solar]

yes

[prisma sundial]

gg everyone, gl next year 🙂

[humble copper]

||im simulating right, but its not stopping after the expected amount of steps||...

[earnest forge]

Make sure you ||remember to wrap around the edges||

[humble copper]

i am

i checked the sample

the individual steps look fine

but its not detecting when none move

[earnest forge]

If you're ||comparing lists||, consider ||just counting the moves you make manually|| instead

Personally I ||wrote a function which does the move, and counts how many things it moved|| which is probably a similar idea to what everyone else did

[tulip elbow]

Merry Christmas!

[earnest forge]

Merry Halloween Christmas!

[tulip elbow]

lolol

[humble copper]

nevermind i was checking the wrong input

ooops

so im assuming as the time between moving everything reduces, the grid gets closer and closer to solving?

[woven sable]

Yayyy

Now all that's left is D23 and D25

[sinful pawn]

still waiting for 3DS

[earnest forge]

But those aren't any of the 5 platonic solids, they'd have to be awkward shapes

[woven sable]

hm?

[earnest forge]

D4, D6, D8, D12, D20

Tbf people have made a D3, which is basically just a triangular prism

(and a D2, otherwise known as a coin)

[earnest forge]

(https://en.wikipedia.org/wiki/Dice_notation)

[woven sable]

similar to Day 22

or was it Day 21

the Dirac Dice

[earnest forge]

Hah

Yes, d(split universe)

[woven oriole]

:D

[hollow wharf]

Merry 🎄🎄⛄

Christmas

[humble copper]

gg

it's been a hell of a ride

[woven oriole]

woah

[humble copper]

thats cool

[crystal shard]

hell yea

[balmy dome]

What does this look like

[crystal shard]

what is it

[woven oriole]

trench

[balmy dome]

French?

Lol

[woven oriole]

depth of sea

we are at the deepest point

hehe

[balmy dome]

😆

So deep lol

[pearl moon]

Is there a non brute force solution to 25

[woven solar]

it's fine to do it the naive way - remember that (this is a minor spoiler for day25 part2 of every AoC year) ||there's no part2, so the part1 is all there is||

[woven oriole]

why will you call it brute force in the first place

👀 ||its just quadratic solution||

[mint hornet]

Wow, this is beautiful

[sonic silo]

pretty lights

[hollow wharf]

omg the size of the input

[hollow wharf]

spoiler!

[woven oriole]

||it made me question myself if my 2d lists simulation will finish on time but yes it does ||

[minor hamlet]

Show off 😉

[hollow wharf]

let's gooooooooooo

[minor hamlet]

I gave up aoc on day 9

got too hard

for me

[woven oriole]

you can always reattempt it again in your life ™️

[minor hamlet]

its just so confusing

I don't understand how ppl are able to do this so easily

[hollow wharf]

ooooh I know what the calendar is now

[hollow wharf]

it is very hard haha

[proud cosmos]

Happy life

Wouldn't call it easy yeah

for some days I invested my entire day essentially

but it feels so good

[muted aurora]

Please tell me today's was not another head scratcher

I just want my 50 stars but I'm done thinking

[proud cosmos]

It was easy

I had some stupid issue I didn't get for a while but you can probably do it in like a few minutes

[muted aurora]

Can I just brute force it?

Ah right no part 2

[proud cosmos]

Yes

[muted aurora]

Cool

Thanks

[rancid creek]

You got it I presume?

[muted aurora]

yeah, just went to make some breakfast first, petty soft

solution takes forever, but we did it boys:

[frigid mural]

🎉

how long did ur solution take?

mine was second-guessingly slow

[muted aurora]

20 seconds lol

[frigid mural]

5:49 :) cos i couldnt be bothered to make it efficient

[muted aurora]

couldn't be bothered to check if I would overwrite stuff, so just making multiple copies every time

[proud cosmos]

wow mine is acutally decently fast then

~2s

lol

[rancid creek]

Yeah, mine was instant too.

[wooden jewel]

I am in spain but the s is silent

[steel bear]

4 more puzzles to go for me !!

[muted aurora]

almost there!

[muted aurora]

oddly satisfying to watch

[maiden brook]

reminds me of snow on christmas morning

[steel bear]

I am seeing a pattern but couldn't quite describe it

[maiden brook]

or just really bad traffic

[muted aurora]

going to post a longer version, just compressing it

screen recordings on my computer are hugee

[woven oriole]

oh i should do this with carriage return

[muted aurora]

I'm just using an ANSI code to just to the top

no fancy library

while True:
  print(chr(27) + "[2J")
  # perform step
  print_grid(cur)
  os.sleep(0.1)

[devout tusk]

#advent-of-code-spoilers-archive message

[mossy stream]

That's super fancy

[muted aurora]

ah shit haven't looked at reddit yet

[humble copper]

thats cool

btw merry christmas everyone

and happy holidays

[maiden brook]

merry christmas 😄

see you all next year
(jk I'll be around)

[steel bear]

Merry Christmas 🌲

[proud cosmos]

Merry hristmas

[iron reef]

This was strangely easy

The brute force is so fast

[sleek cave]

day 25, finally a not-so-hard puzzle :D

anyways merry Christmas guys!!!

what an event

absolutely amazing

(even though I basically couldn't do anything after day 18 except for day 20 lmao)

I might reattempt the harder puzzles when I have time... waiting for that one day when I'll finally get 50 stars lol

also @woven oriole congrats! our first cat dev completionist

[frigid mural]

Merry Christmas to all of you, 'twas a fun ride with some really great people. I'll see y'all around :) 👋

[maiden brook]

so long, and thanks for all the lanternfish

[frigid mural]

its kinda sad theres only 25 days of advent

[steel bear]

Yeah ikr

[woven oriole]

lmao thanks , merry christmas 🎄

started out with numpy 👀

i shall too lets go

[woven oriole]

day 26 : lanternfishes occupied mars because no space

[wooden jewel]

I said previous weekend "I think we had the worst part, should be relatively smooth sailing from here on out"

Oh how young and naive I was

[unreal willow]

could someone explain?

pls

[woven sable]

Well, they all move simultaneously so when the sea cucumber you circled in the first step tries to move and wrap around, it takes that state meaning that the other sea cucumber is blocking it so it doesn't move

[woven oriole]

it was intended to go there but there's already one v in its place which moves downward since its dot below it

[unreal willow]

thanks, now I get it

[woven sable]

Np!

[ancient geyser]

Data Analyst or Data Engineer ?

[wooden jewel]

Data Lanternfish

[ancient geyser]

hahaha

Lanternfish?

[wooden jewel]

(I'm joking, you're in the channel for advent of code, which is a yearly event around this time)

Maybe try #career-advice ?

[sleek cave]

numpy. is. amazing.

[proud cosmos]

Nice!

[sleek cave]

all the times when i did sum(bruh < something for seq in 2dseq for bruh in seq), i could've just numpy'd lmao

[proud cosmos]

To be fair that is not relaly that much worse in my opinion

but it is nice it works

[sleek cave]

it's not, but when the entire aoc is just 2d and 3d matrix puzzles (\s, but these puzzles are pretty common), it gets pretty bad

so many early days can be solved elegantly with a few lines of numpy

the later days, not so much, but the early days with numpy is :chefskiss:

YOU CAN EVEN SPECIFY AXES AND CUMULATIVE SUMS?!!

i'm in love halp

[wooden jewel]

don't tell this guy about array programming languages :P

[sonic silo]

dont tell him about matlab or R or julia, or he goin crazy

[wooden jewel]

julia is an apl

can't change my mind

[candid sigil]

For me nothing beats R's "most things are vectors or functions" approach. Simple, yet versatile and convenient to work with. But NumPy's not far behind.

[sonic silo]

¯_(ツ)_/¯

maybe one day i'll revisit R

i just touched that lang ever so slightly during my statistics class in my first year of college

[candid sigil]

It has always felt more natural than Python for data manipulation to me, although I'm biased in R's favor because I learned it first. It does have more quirks and questionable implementation decisions than Python, though.

[sonic silo]

that's saying a lot about R's implementation lmao

[candid sigil]

I don't think it has any truly awful features, but I would get rid of stuff like partial argument matching if I had the power. There are several elements like that that feel like relics of a time when R was mostly used interactively or for small analysis projects.

[hollow wharf]

kinda confused, i just randomly decided to program the final day
i'm done programming it but I'm confused what the input is, it came from the previous day didnt it? does anyone have it?

[humble copper]

wdym

day 25?

the input is the same as always

if you mean part 2 then ||you need to fully complete all other days in the calendar to finish part 2||

[hollow wharf]

it's my first day trying it so i dont know what it usually is

oh its at the bottom

[woven sable]

Until when do we have time to complete the 50 stars to get the AOC Finisher role?

[wooden jewel]

Is that automatically applied?

[hollow wharf]

It will be automatically applied and will run until February iirc

[sonic silo]

do i not get the role

[humble copper]

january i believe

yeah if you get all 50 by january you get the aoc completionist role

so that gives you roughly 5-6 days

[sleek cave]

ah, well then i'm basically dead. it's been a good game, i'll see y'all next year :D

[humble copper]

well thats just if you want the role

[sleek cave]

although i am learning numpy now and it's absolutely amazing

[sleek cave]

i know

i won't be able to get it, these puzzles just get way too hard for me (for now)

next year i'll see if i can beat this year's score

[narrow crown]

😏 I'm already at day 4

definitely will be able to finish the rest before january

[narrow crown]

Where can I learn NumPy from?

[humble copper]

youtube tutorials might be useful

[narrow crown]

ty

[hollow wharf]

famous last words

[sleek cave]

https://numpy.org/doc/stable/user/whatisnumpy.html starting from here and reading through the chain of docs is what i did, but if you're more of a video-learner you can find some tutorials on youtube

[narrow crown]

:D thank you

[narrow crown]

I've just given up XD

[sleek cave]

you're welcome!

[sleek cave]

same lol, trying today because it seems easier

day 19, 21, 22, 23, and 24 are just :sigh:

[hollow wharf]

We did all the 50 stars without any external libs

That's very much doable without numpy

[sleek cave]

amazing

[narrow crown]

Me who doesn't know many useful external libraries:

[sleek cave]

ah yes, input49.py
imposter spotted

[proud cosmos]

ah yes

i

input49.py is better tho

sad

[sonic silo]

i

[sleek cave]

what a ride

welp, i may not have saved christmas, but i got 40 of them shiny stars lmao

double my initial goal

this has been such a fun event, i absolutely love aoc

will come back to the harder puzzles later to see if i can get 50 stars lol

[muted aurora]

@sleek cave ayy, at least you got your 40! nothing to scoff at

some of the later ones were very hard

[sleek cave]

i know, i'm extremely glad lol. aoc is probably one of the coolest and most fun programming events i've ever participated in, even though the harder puzzles really destroyed my brain lmao

taught me so much stuff

absolutely amazing

[candid sigil]

I got 36. More I think than last year. Not sure how much further I'll go, though I would like to finish off part 2 of day 21.

[narrow crown]

i got 8 B)

[sleek cave]

d21p1: bruh so ez
d22p2: now, become doctor strange.

yeah, I didn't even understand part 2 lmao

universe trifurcation, can't be me

[candid sigil]

I spent a few hours but gave up when my computed answer for the test input was off by a mere 7 orders of magnitude.

I knew from the reference to Dirac that ||quantum nonsense|| would be in part 2, but nothing like that.

[ruby epoch]

12 ⭐️ here

[candid berry]

Day 22 took me a while, but day 19 absolutely wrecked me

That took me 9 straight hours, by far the worst I did on any of them

My second worst was 5 hours for day 23

[sleek cave]

worst: day 22, 6+ hours unsuccessful attempt
2nd worst: day 18, 5.5 hours lmao

others are unattempted

[woven solar]

@Advent of Code Good morning! Day 26 is ready to be attempted. View it online now at https://adventofcode.com/2021/day/26. Good luck!

[steel bear]

wait wdym by day 26 , isn't it until day 25

[crystal shard]

it is meant to be a joke

[steel bear]

lol 🙂

[potent pumice]

lol

sir-lancebot#992

[hidden copper]

GitHub

Advent of code ends before December ends

[woven oriole]

lmao

[hybrid schooner]

well at least i tried

xd

[robust heart]

I'm taking AOC at my own pace imo

As long as I can find a solution

I just hope I can ask for questions even after December ends

Is #algos-and-data-structs appropriate?

[woven oriole]

you can always assign a help channel

[ruby epoch]

You got slightly further than I did. I finished day 6, but didn’t start day 7 yet

[sleek cave]

lmao nice

I'm so inspired by this event, I want to make my own aoc called "are you smarter than a (snailfish) fifth grader," which would be basically 14 puzzles released over a week that only regard snailfish math

and each day is a higher grade ending at gr5

[hollow wharf]

nO

[sleek cave]

lmao why not

[hollow wharf]

We still have nightmares of that day

[sleek cave]

snailfish multiplication :evil:

[sleek cave]

I plan for the actual aoc puzzle to come at day 6 (with slight modifications), so it won't get much harder (because I can't make hard puzzles lmao)

but this is a project once I get more experienced. will probably start working on it in 2 to 3 months after I've done more aoc puzzles from past years

[sharp panther]

dew it

[hollow wharf]

yy

hw are you doing

[muted aurora]

If you're seriously considering this hmu I can help make puzzles

[sleek cave]

damn right, i'm seriously considering this 😄
with help from more experienced people, we'd be able to make harder puzzles :DDD

i'd greatly appreciate you collaborating!

[sleek cave]

wanna help? it'd be great if we had a team, maybe we could even call the other cat devs

[woven sable]

I'd love to help too ^^

[sleek cave]

i see fronto typing furiously
too late he already sent message lmao

[woven sable]

Lmao

[sharp panther]

sounds fun

[sleek cave]

lessssgo!! perhaps if we get enough people and enough puzzles, we could ask event team to make this an actual event?

okay let's not get too ahead of ourselves lmao

[muted aurora]

Probably just do a trial run first to see how it goes 😅

[sleek cave]

yeah, my bad, got too hyped

[woven sable]

But if we did make puzzles and what not, it should be its own thing and have a totally different concept than AOC

[sleek cave]

i've been thinking about this ever since day 18 released

[sleek cave]

of course

[woven sable]

Did doing the homework for snailfish inspire you lol

[sleek cave]

i was simply inspired by snailfish lol

[sleek cave]

yep

if y'all want, maybe we can discuss details in a private server/group chat?

[sleek cave]

@muted aurora @woven sable @sharp panther sent y'all links!

[proud cosmos]

Nice

Making puzzles is hard otherwise I'd love to help

But I'm pretty bad

[muted aurora]

On the contrary, I'm terrible at organizing anything, but making puzzles and programming challenges I can do

[minor pike]

me too :)

[river hollow]

I don't think the .aoc leaderboard commands knows my AoC user name, and can't place me on the leaderboard. Is there a way to link it?

[hidden musk]

yep

uhh, let me find the command. i think it's .aoc link

[river hollow]

Thanks!

\o/ that worked

[hollow wharf]

with open("Day3\input.txt") as f:
    lines = f.readlines()
    lines = [line.replace("\n","") for line in lines]


def common(list,index,rule):
    ones = zeros = 0


    for bin in list:
        if bin[index] == "1":
            ones += 1
        else:
            zeros += 1

    if rule == "oxygen":
        return "1" if ones >= zeros else "0"
    else:
        return "0" if zeros <= ones else "1"



def pick(lst,common,index):
    res = []


    for bin in lst:
        if bin[index] == common:
            res.append(bin)

    return res

def get(rule):
    lst = lines.copy()
    pos = 0
    

    while len(lst) > 1:
        
        lst = pick(lst,common(lst,pos,rule),pos)
        pos += 1


    return int(lst[0],2)

oxygen = get("oxygen")
carbon = get("carbon")

print(oxygen * carbon)

yo guys what can i improve on

in this solution

it's for the 2nd part of the 3rd day

https://adventofcode.com/2021/day/3

[digital ember]

bonk, not sure how helpful i would be, but would love to :D

[steel bear]

hey can add me too 🙂

[cyan sleet]

Embrace the niceness that is list comprehension and generator expressions. With it kinda makes it hard to justify not inlining pick.

def pick(lst,common,index):
    return [bin for bin in lst if bin[index] == common]

You could similarly do

ones = sum(bin[index] == "1" for bin in list)
zeros = len(list) - ones

also in your input reading you could do line.strip() rather than the replace.

[candid sigil]

Starting day 16 now. Keeping track of the length of packets is going to be a nightmare.

[muted aurora]

@proud cosmos I'm giving up on trying to optimize my Day 23, can you pls send your code, I'm just going to adapt from that to get the speed

[proud cosmos]

is my day23 that fast?

[muted aurora]

mine's like 22s didn't you say you got it down to like 2s

[proud cosmos]

its 5s atm not sure what I did

I also feel like I had that faster at one point

oh well

[muted aurora]

well either way still significantly faster than mine

and idk what I'm breaking with dijkstras

oh wait right your day 23 didn't work properly did it

[proud cosmos]

it should now

it was me having weird seen

this is my code atm

I should remove the prints ig that would help

also heuristic is just 0 atm.

i still haven't found one that really worked..

I'm confused now why is my "is_finished" 4 permanently

that looks wrong

It gives the right result, weird.

oh that's just that one first check ig that's fine

Not sure why it being set to amount breaks it though this makes no sense sometimes.

so my is finished check being correct breaks stuff haha..

Yeah my day 23 code is just pure confusion for me now I am not sure what is happening

I should just rewrite this.

(it was heuristic not being 0 that broke it oops.)

my heuristic almost gives right result but not quite

:(

would be so good

[muted aurora]

well that seems essentially identical to mine, I meant more like the actual code to find the moves and stuff

[proud cosmos]

I see

I'll just give you everythig just uh

its a mess

I can definitely make this better..

I'd assume the worst thing of your code is you using objects for the states right?

or am I misremembering

cause my stuff is just strings

[muted aurora]

even so

[proud cosmos]

https://paste.pythondiscord.com/liyisucezi.sql

My get possible moves function really isn't good

[muted aurora]

same issue for my C implementation, where I would suspect there's not much ovrhead since I'm hashing manually

I feel like there's some algorithmic thing I'm messing up

[proud cosmos]

Yeah well.. I feel kinda bad for even showing you that code

it's a lot of things that have been created cause of other reasons but never fixed

like me going through rooms in reverse order is what I did cause I used to have this as a generator and that seemed smarter (?) now kinda useless especially since I search everything atm..

Also didn't know this was the non spoiler chat

oops.

Yeah I need to rewrite this..

[rancid quarry]

lines = input.read().rstrip().splitlines()
cucumbers = [[] for x in range(len(lines))]

i = 0
while i < len(lines): 
    for line in lines:
        for c in line:
            cucumbers[i].append(c)
        i += 1

#above can be ignored
k = 0
j = 0

cucumbers2 = [[]]

for line in cucumbers:
    for cucumber in line:
        cucumbers2 = move_east(cucumbers, k, j)
        j += 1
    j = 0
    k += 1

can someone help me? I really dont understand how when i assign cucumbers2 to the result of move_east(), the list cucumbers changes itself?

its not like im assigning cucumbers2 = cucumbers, im only assigning it to a result of my function so i dont get how cucumbers2 and cucumbers are related

[proud cosmos]

what does move_east do

(as in show the function)

it probably manipulates the original list

[rancid quarry]

thats what i thought too, but then i changed the variable name and it still changed both lists

def move_east(cucumbs, pos1, pos2):
    if cucumbs[pos1][pos2] == ">":
        if pos2 == len(cucumbs[pos1]) - 1 and cucumbs[pos1][0] == ".":
            cucumbs[pos1][0] = ">"
            cucumbs[pos1][pos2] = "."
        elif pos2 != len(cucumbs[pos1]) - 1:
            if cucumbs[pos1][pos2 + 1] == ".":
                cucumbs[pos1][pos2 + 1] = ">"
                cucumbs[pos1][pos2] = "."
   
    return cucumbs

its messy code but im just trying to get it to work

[cyan sleet]

variable name doesn't matter

[proud cosmos]

Yea

[cyan sleet]

it's still referring to the same list

[proud cosmos]

^

make a copy in move_east first, or similar

For 2d lists a bit annoying might be nicer to just change the solution to always use the new list

[rancid quarry]

def move_east(cucumbers, pos1, pos2):
    cucumbs = copy.copy(cucumbers)
    if cucumbs[pos1][pos2] == ">":
        if pos2 == len(cucumbs[pos1]) - 1 and cucumbs[pos1][0] == ".":
            cucumbs[pos1][0] = ">"
            cucumbs[pos1][pos2] = "."
        elif pos2 != len(cucumbs[pos1]) - 1:
            if cucumbs[pos1][pos2 + 1] == ".":
                cucumbs[pos1][pos2 + 1] = ">"
                cucumbs[pos1][pos2] = "."
   
    return cucumbs

``` so like this?

[cyan sleet]

does copy do a deep copy?

I've never bothered with it

[placid lake]

nope, use copy.deepcopy

[cyan sleet]

I would personally do something like

cucumbs = [row[:] for row in cucumbers]

[rancid quarry]

ok thanks

[candid sigil]

Giving up for now. I just can't think of an algorithm that would work.

[proud cosmos]

for?

[livid gull]

advent of code 2015 almost done

[steel bear]

I am at Day 23 almost getting it 🙂

[sleek cave]

@digital ember @steel bear sent y'all links!

[normal umbra]

Hey there. I am trying to solve day 4 (got a bit late lol). So solving part 1 was very easy, but I am stuck on part 2. The final answer I am getting is too high. Anyone got any tips or something? Nvm got it

[late linden]

@woven solar

define(`strlength', `ifelse($1, `', 0, `eval(1 + strlength(substr($1, 1)))')')

strlength(foo)

this is m4

a macro language like the C preprocessor.

this is a recursive macro that calculates the length of a string.

look on in horror.

OH SHIT

THIS IS THE ADVENT OF CODE CHANNEL SORRY

[sonic silo]

man

i wanna get done with all my work for this christmas

so that i can begin solving aoc puzzles from prev years properly agian

[muted aurora]

bruh I just sat down and did days 1-18 in one sitting

2015 so ez compared to this year lol

[woven sable]

https://m.webtoons.com/en/challenge/advent-of-code-2021/hydrothermal-venture/viewer?title_no=713188&episode_no=6
This is amazing

[livid gull]

only one i had to do research on was day 12 the JSON one, day 7 was my favourite how did you code it cus mine seems really inefficient but i like it ||https://pastebin.com/GkVCJF1d||

[muted aurora]

Since each wire could only get a signal from a single source, I did it backwards

started with a, and tried to recursively compute the things it was dependent on

that way I don't need to worry about any extra wires that don't do anything

[livid gull]

i just kept track of the registers?wires? that had an integer value and then if another one used them i could then add that to the ints and iterate over them

[muted aurora]

I tried doing that too but since each of the things could be integers/wires it got a bit annoying the way I was doing it

and I was getting way too many wires when printing all the values out to debug

so doing it backwards helped with that

didn't spend too much time trying to think of a decent way tbh, just wanted to get stuff working

all my days are just basically hacked together LOL

I cheesed day 8 so hard lol, just used eval() and repr()

[livid gull]

that day was so hacked lol, my day 7 was https://pastebin.com/4zYKQkLy

[livid gull]

i used eval for p1, and for p2 used count, got it to 2 one liners for each part

[muted aurora]

count?

[livid gull]

s.count("\") + s.count('"')

2 slashes

i dont know why discord keeps them hidden

[muted aurora]

ah lol that's smart

[livid gull]

...

how do i spoiler code?

||print(sum(len(s) - len(eval(s)) for s in open("Day8.txt").read().split("\n")))
print(sum(s.count('\') + s.count('"') + 2 for s in open("Day8.txt").read().split("\n")))
||

[sinful pawn]

I don't think you can

[livid gull]

i guess you get no syntax highlighting then lol

[sinful pawn]

but this is why we have #advent-of-code-spoilers-archive

[muted aurora]

this is for 2015, I don't think many people are trying to do it I guess

[sinful pawn]

oh

2015

[hollow wharf]

i know a good code

=SUM(A10:A15)

[wispy glacier]

WOAH! i know this is an old message but it's what i last read in this channel and that is so pretty 🎄

[north mist]

what is the advent of code

[narrow crown]

https://adventofcode.com :O

pain and suffering for a person like me :P

[delicate dirge]

found the people at the top of the global leaderboard

[peak dock]

I have completed aoc 2021!

[hidden musk]

🎉

[woven sable]

Yay, nice job!

[narrow crown]

I have completeted aoc day 4!

[woven sable]

Nice! that day was a tough one

[narrow crown]

lol ive about given up on aoc this year now

my coding skills are not nearly that good

[woven sable]

well, this can be a learning experience for them if you want it to

[narrow crown]

lol true

instead ive been working on this other program

[pallid shoal]

Hi

[rose fable]

I completely forgot about this until now but I figure I'll do it as an exercise

[steel bear]

this inspires me to do AOC 15 rn HAHAHA

[muted aurora]

Pretty fun isn't it?

[steel bear]

oooh haven't done yet 😅

[muted aurora]

Try it if you have some time to kill, the first 15-16 days are all pretty easy

[radiant grail]

Can anyone please explain how to use advent of code

[woven sable]

You create an account or sign in with something and then you just go to each day of the year you want to go to, get your input, the answer in the text box based off your input

[radiant grail]

Is that it?

[woven sable]

Yep

[main edge]

[hollow wharf]

how can i export the json data to a txt without 1 thing

[hollow wharf]

get this error when i wanna install psutil

why?

[steel bear]

the error shows your microsoft visual c++ is not updated, maybe try updating it ?

[hollow wharf]

its updated

[steel bear]

what version is your Microsoft Visual C++, the error message says 14.0 and above is required

[hollow wharf]

why c++

its python

i thout u told me vsc

[steel bear]

yes, its in the error message

[hollow wharf]

ok dude ty

[steel bear]

Windows Python needs Visual C++ libraries installed via the SDK to build code, such as via setuptools.extension.Extension or numpy.distutils.core.Extension. For example, building f2py modules in Windows with Python requires Visual C++ SDK as installed above. On Linux and Mac, the C++ libraries are installed with the compiler.

https://www.scivision.dev/python-windows-visual-c-14-required/, just did a google search 🙂 🙂

[hollow wharf]

ty dude 🙂

[normal umbra]

Hello. Can someone help me understand day 7 problem? I don't even understand what I have to do 😕

[cyan sleet]

You're given a bunch of x positions of crab submarines, you are given how much it costs for them to move along the x-axis, what's the minimum cost to get all to the same x position?

[normal umbra]

Alright so I understood that, but then I got confused by this example. Specifically, the three 2s in there

[cyan sleet]

three of them are already at x=2

[normal umbra]

oh

so there can already be more than one at one point

there position in the list has nothing to do with the fuel cost right?

[cyan sleet]

it's one line per x value in the list above

no

[normal umbra]

Alright thank you

[livid gull]

if i need a sorted stack what library do i need to use?

[proud cosmos]

Depends on what exactly you mean but probably something like a heap

not exactly sure how sorted and stack work together

tbh

[livid gull]

idk data structures lol, what is normally used for dijkstras

[velvet stag]

heapq

[sleek cave]

a priority queue, thus the heapq library

[livid gull]

ty both

well thats reduced my time from 8 seconds down to 1.8 seconds

[normal umbra]

For day 8 pt 2, do I have to like, read the sequence of strings on the left of the | and figure out what the pattern of the abcd is? Like its different for each line?

[ruby epoch]

I'm working on the same task right now, and yea, each line is different

[normal umbra]

Ah fuck lmao

Thanks for the confirmation

[ruby epoch]

Join me in #advent-of-code-spoilers-archive to brainstorm how to figure this out? I have some ideas but I'm stumped

[normal umbra]

I was just giving it a read, altho I am going to have dinner rn. I will start solving it a few hours later. Mind if I ping you then?

[ruby epoch]

Sure

[normal umbra]

Sweet

[lost dust]

Complete noob, i was wondering if anyone could guide me with the seconds puzzle. I think i need to use array slicing but not sure

Do i need to delete /n from the text file

[proud cosmos]

\n is a "newline"

I don't think it's part of the actual input in any of the days this time

so yeah you can just remove it

(well, or split at it, cause usually every line means something else)

[lost dust]

how can can I remove or split it? @proud cosmos thank you

[proud cosmos]

Strings have a .split method that does just that, removing probably with .replace