advent-of-code

[compact wasp]

lmao

[magic oxide]

lmao

did any of you read that or do I have to send it again

[glad wigeon]

legend

[earnest forge]

Congrats!!! It's way harder to get LB now than it was when I started, so that's really cool!

[compact wasp]

oliver still gapping me though 😦

[magic oxide]

Oliver insane

[compact wasp]

he got lb on d1p2

and i didn't

so

[magic oxide]

[mossy basin]

i think i spent the longest on the parse

[magic scaffold]

re go boom

[magic oxide]

who even is oliver

[rancid creek]

This should have been the day 1 puzzle. 🙂

[earnest forge]

^

requiring a parsing change for part 2 is nasty and definitely not day 1 material

[covert arrow]

||(\d+) (\w+)|| isn't too bad tbh

[earnest forge]

I mean tbth I just hardcoded in the specific cases

That might be useful for the golf though

[rancid creek]

people changed their parse on this one?

[earnest forge]

No for day 1

[rancid creek]

I screwed up by not noticing the 13 changed to a 12 for the input data

[earnest forge]

Ooops lmao

[humble copper]

tyyyy!!! this is like my 5th year doing these (3rd year competitively), so i'm really happy to have finally managed to do it

[compact wasp]

congrats

[magic oxide]

nice

[rancid creek]

It's amazing how little a difference there is between top 100 and top 2000. Getting to the top is really going to be a challenge for anyone not pushing for it. Well done.

[modern terrace]

at least i didn't do bad in p2 this time

[cyan sleet]

I have managed to read a line

progress

[unique osprey]

wow, man i got held up on p1 for no reason

[gentle cape]

I spent abt 10 min on part one bro i need practice

[rancid creek]

was it the 12/13 change?

[gentle cape]

Of course, I am attempting everything in Haskell

[unique osprey]

||i was summing all the cubes per game, instead of accounting for each set individually||

[compact wasp]

how much difference is there?

i'm curious

there was like a 3 minute difference between 100 and 300 for day 1 part 2

which is a lot

[magic oxide]

day 1 part 2 was an anomaly tho

[compact wasp]

true

[magic oxide]

today's back to normal difficulty

[compact wasp]

thankfully i didn't throw

[magic oxide]

now go golf

[stable shale]

I spent an extra 10 minutes on Part 1, because I ||accidentally only got the first digit of the game ID||...
AAAAAAA

[gentle cape]

Is it fine if I skip a few days? I’m on vacation for this weekend

Like, are they required for the next puzzles?

[stable shale]

No, previous day's aren't required

[gentle cape]

Thanks!

[candid berry]

it's ok to skip problems, it just won't be great for your score 😄

[zinc kindle]

Did it again in Rust. I suspect I would have done it a lot faster had I access to python's ||collections.Counter||. They have ||>= and | || which would have trivialized a bunch of code I spent writing. Eh, made it anyway.

[gentle cape]

[compact wasp]

fr i skipped the first few days last year and i got completely screwed on this server's leaderboards

[gentle cape]

Rust 👌

[placid lake]

hmm why was today easier than yesterday

[gentle cape]

Ong though

I mean

Part one yesterday was easy

[zinc kindle]

Usually not required, but there are some occasions where solution for one problem builds on previous one (e.g. 2019 intcode) or is an extention of a previous day's. I'm going to be in the same situation and intend to fulfil the backlog on my own time.

[candid berry]

honestly I'm still mad about yesterday's examples.

[compact wasp]

well they are required for day 25 part 2

[gentle cape]

Oh…

[compact wasp]

yeah

[gentle cape]

Dang

[candid berry]

well, yes, but that's not really a puzzle

[gentle cape]

Time to speedrun next week

[candid berry]

that's just a reward for having completed all the other 49 puzzles

[compact wasp]

fair enough

[magic oxide]

well day 25 2nd star

[modern terrace]

the 9 minutes delay was horrifying

[candid berry]

9 minutes delay?

[modern terrace]

the time between day 2 starting and me noticing

[candid berry]

oh lol

need alarms 😄

[modern terrace]

true

[compact wasp]

yeah completely locked you out of both boards 😭

[placid lake]

ballmer peak was not necessary, but it's a friday night :D

[zinc kindle]

You don't have to do them in order, or complete them the day they're released. I'll be happy if I get all 50 stars by the end of La Befana (12 days after Christmas) local time.

[modern terrace]

i got 20 and 25 minutes though

so it's not like the 9 minute delay was significant

since just solving p1 took me 11

[compact wasp]

😔

[candid berry]

class 'cereal' (slow)

😛

[earnest forge]

LOL

[modern terrace]

but-

but that's supposed to be runtime, not coding time ;-;

anyway golfing time

[naive stone]

why do i always make the odd decision to write tons of bugs into my code

[topaz coyote]

Some of us have coding as a hobby, and some of us havee debuging as a hobby.

[royal gull]

It's like sculpting, you remove everything that isn't the thing you're trying to sculpt. 🙂

[naive stone]

heh

[jagged hare]

right there with you 🍻
(I started a beer about 20 min before the puzzle dropped)

[candid berry]

I made a G&T after finishing it 🙂

[jagged hare]

nice

[pearl glen]

skill issues

[naive stone]

you're not wrong

[hallow birch]

Man that was way easier than day 1 lol

[pearl glen]

while True:
print('I'm brainfucked')

[hallow birch]

I had some fun w/ sets

[earnest forge]

I think you mean +[I'm not writing the code here it would be way too long]

[pearl glen]

😂 😂

yea brainfuck actually brainfucks you

[zinc kindle]

Yeah, today was easier than yesterday's part2. A good chunk of my time was spent on input parsing I think (wonder if I should bite the bullet and learn regex).

[hallow birch]

You converted me lol, it's pretty nice

[earnest forge]

🙃

[hallow birch]

I like the separate function for parsing input

[earnest forge]

I mean the best part is clearly the auto tests lol

[hallow birch]

I mean not really, you can just have assert result is <answer> lol, had that before

[hollow summit]

salt die has competition

oh wait nvm he wrote both???

[earnest forge]

He co-wrote early versions of aoc_helper, but I've been maintaining it solo for like 2 years lol

[earnest forge]

Yeah but auto tests, it pulls them out of the webpage

[hallow birch]

Yeah tru

Btw is there a function for reading the prompt in the terminal?

[earnest forge]

Not currently

[hallow birch]

gh issue time lol

[earnest forge]

Lmao

It's totally doable but I like viewing it in a browser because you can user-style the important bits of the page

(I did that, they now glow yellow instead of white)

[cerulean epoch]

imo yday's part 2 wasn't as much hard as much as it was poorly written; if they had emphasized in the text that, for example, EIGHTHREE should return 83, then that would've cleared up a lot of confusion

[shut trellis]

Wrong Channel

You are not allowed to use that command here. Please use the #aoc-bot-commands, #sir-lancebot-playground, #bot-commands channel(s) instead.

[cyan sleet]

part 1 done 🥴

god I'm so terrible at functional languages

[lusty rover]

In OCaml?

I quit

I did the Python, I'll do language learning.... sometime that's not 2am

[cyan sleet]

ok, I spent way too much time cleaning up code rather than just solving part 2

in any case, done

now to clean up this absolute mess...

[halcyon pulsar]

What is the name of the private leaderboard?

[somber trench]

why is part 2 of day 1 ridiculously hard

[magic oxide]

it was an interesting decision for sure

[somber trench]

it says my input was too high, i cant figure out what went wrong since im going through it now, still no errors in gap of judgement

[modern terrace]

have you tested these edge cases ||```
twoneight
nineighttwo
5eightwo

`28`, `92`, and `52` and totals to `172`||

[somber trench]

im using regex so it should detect it fine

ill test it tho still

[somber trench]

it seems to not be able to detect those type of things, maybe thats the issue

idk how to fix it tho

[modern terrace]

1. do not use .findall()

.finditer() too

[somber trench]

what would u suggest then? cause idk what other thing would help with keeping them in the order they appear?

[modern terrace]

continue in #aoc-solution-hints

[inner bough]

I have a solution to puzzle 1 written in brainfuck. Any recommendations on where to post it?

[magic oxide]

the spoiler thread works

we don't really care about the language

[somber trench]

day 2 is confusing how it is written

[modern terrace]

too much words, aye?

[somber trench]

its hard to understand it

[versed plaza]

how can i connect my discord to aoc
so that i can get that fancy colored name in here?

[modern terrace]

for example ```
the ID of the game
|
| set 1 set 3
V _____________ _______
Game 1: 3 blue, 4 red; 1 red, 2 green, 6 blue; 2 green
----------------------
set 2

each set is separated by a semicolon (;)
each set has a specific format, example for set 1...

number of cubes...
| ...blue cubes
V vvvv
3 blue, 4 red
^ ^^^
number of red cubes

with each type of cube (red, green, or blue) distinguished with a comma (,)

[somber trench]

this is still super confusing, might be cause im sleep deprived, 4am with no sleep go brrrr

[modern terrace]

have to find the games that can use only 12 red cubes, 13 green cubes, and 14 blue cubes

[somber trench]

each game format is different though with each set, idk how i would manage to use that

[modern terrace]

||so basically check each color of cube in that set and the compare that number for its limit ```
game 1 is okay because...

game ID
 |       set 1                              set 3
 V  _____________                          _______

Game 1: 3 blue, 4 red; 1 red, 2 green, 6 blue; 2 green
----------------------
set 2

set 1:

3 blue, 4 red
^ ^^^^
the limit for blue cubes is 14
3 < 14, so that's okay

3 blue, 4 red
^ ^^^
the limit for red cubes is 12
4 < 12, so that's okay

set 2:

1 red, 2 green, 6 blue
^ ^^^
1 < 12, okay

1 red, 2 green, 6 blue
^ ^^^^^
the limit for green cubes is 13
2 < 13, okay

1 red, 2 green, 6 blue
^ ^^^^
6 < 14, okay

and so on for set 3...
each game is filtered like that

[somber trench]

|| so i need to check if the cube amount is under the amount allowed and if it meets that for all sets then its a possible game? ||

[modern terrace]

||yes||

[somber trench]

gonna be a pain

[late portal]

guys i don't understand aoc puzzles, what should i practice bforehand to solve them?

[hollow summit]

reading

[late portal]

you mean i improve my vocab?

[hollow summit]

sure? whatever helps you understand the problem statement better

[late portal]

its puzzles are weird tbh

[hollow summit]

you're working for Santa

don't complain

[late portal]

hmm no i guess

[torn sapphire]

AOC questions are usually purposefully pretty multi-disciplinary, but there's a bit of a focus particularly on DSA stuff and the parsing of input (e.g. today's challenge has a big focus on parsing input textual data into a more useful representation)

[late portal]

whats DSA?

[torn sapphire]

data structures and algorithms

[late portal]

oh

[torn sapphire]

i suppose there's a bit of overlap with the things AoC asks of you and the things that a competitive programming contest would ask of you

[late portal]

i guess i need to practise for 1 year more

[torn sapphire]

there's always previous years' questions to practice on too, especially the ones on the first few days of december

[oak pollen]

For Day 2, Part 1, is it just if total Game's red > 12, total blue > 13, total green > 14 then it's impossible?

[late portal]

no , i mean ill practise python concepts more

[torn sapphire]

yeah, if at any point more than those limits are pulled out at once, then that game wouldn't have been possible

[oak pollen]

okay because in the sample info they split it into sets and that confused me lol

thank you

[white laurel]

yes each set should not exced these numbers

not the sum of colors of all game

[oak pollen]

how do i know what to split into sets then?

what determines a set?

[white laurel]

by >> ;

[torn sapphire]

the sets are delimited using ;s

[oak pollen]

oh god i didnt even see that

thank you so much

[torn sapphire]

in the context of the sorta AOC world, its like in the example game 1:||

• The elf grabs some cubes, and shows them to you: 3 blue cubes, and 4 red cubes
• The elf puts the cubes back in the bag, and then shows you 1 red cube, 2 greens, and 6 blues
• The elf throws the cubes back in again, and pulls out their final hand: 2 green cubes||

[oak pollen]

👍

[hollow wharf]

funny how in day 2 on part 1 i spent like 30mins but on part 2 it was not even 5mins 😭

[rapid wolf]

parsing problems do be like that

[hollow wharf]

i should learn regex

for that

[lost nest]

guis

is this good code?

def is_game_possbile(input: str):
    input = input.split(';')
    input[0] = input[0].split(":")[1].lstrip()
    red_array = [] 
    green_array = []
    blue_array = []

    for i in input:
        for j in i.split(','):
            if "red" in j:
                red_array.append(int(j.split()[0].lstrip().rstrip()))
            elif "green" in j:
                green_array.append(int(j.split()[0].lstrip().rstrip()))
            else:
                blue_array.append(int(j.split()[0].lstrip().rstrip()))

    red_array.sort()
    green_array.sort()
    blue_array.sort()

    return red_array[len(red_array)-1] * green_array[len(green_array) - 1] * blue_array[len(blue_array) - 1]

this thing works correctly

[hollow summit]

your function starts with is_ which should imply that it returns a boolean

but looking at the return statement, i'm not sure if that's the case

[hidden musk]

how fortuitous that i already planned to do ocaml this year and half set it up already

[lost nest]

but ik that this thing works

[lost nest]

i forgor to change the function name though

i edited 2.1's code

[hollow wharf]

.lstrip().rstrip() can be replaced with .strip()

[iron olive]

Just sat down to try my hand at day two, am not even properly understanding the question

Off to a great start

[lost nest]

ahh ok

[astral iron]

What is lstrip and r strip

[hollow wharf]

!d str.lstrip

[marsh current]

str.lstrip

str.lstrip([chars])```
Return a copy of the string with leading characters removed. The *chars* argument is a string specifying the set of characters to be removed. If omitted or `None`, the *chars* argument defaults to removing whitespace. The *chars* argument is not a prefix; rather, all combinations of its values are stripped:

```py
>>> '   spacious   '.lstrip()
'spacious   '
>>> 'www.example.com'.lstrip('cmowz.')
'example.com'
```  See [`str.removeprefix()`](https://docs.python.org/3/library/stdtypes.html#str.removeprefix) for a method that will remove a single prefix string rather than all of a set of characters. For example...

[hollow wharf]

rstrip literally the same but for right side

[atomic cairn]

Does day 2.2 still work with impossible games or is that part not used anymore

[magic oxide]

it's not used

[atomic cairn]

Alright thanks

[astral iron]

Is there the opposite of that

NPM leftpad 😭

[magic oxide]

there's str.rjust, or you could use the f-string format spec

[agile orbit]

Hi! I think theres something wrong with my AoC input for day 1. The task states I need to combine the first and last digit of each line - however, on line 5 of my input I have the string zphgdcznqsm2 which only has 1 digit (2) at the end.

Am I missing something here?

The newly-improved calibration document consists of lines of text; each line originally contained a specific calibration value that the Elves now need to recover. On each line, the calibration value can be found by combining the first digit and the last digit (in that order) to form a single two-digit number.

[rapid wolf]

Well, what's the last item in a one element list.

[covert arrow]

check the last line in the example input (||treb7uchet becomes 77||)

[agile orbit]

Aha thanks, I feel stupid now!

Appreicate the sanity check :)

[mossy trout]

Who has done day 2 advent of code?

[agile orbit]

||```py
sum = 0
with open('day1.txt') as puzzle_file:
for line in puzzle_file:
numbers = [num for num in line.strip() if num.isdigit()]
if len(numbers) == 1:
sum += int(numbers[0])*11
else:
sum += int(''.join(numbers[::len(numbers)-1]))

Done! Anyone done a 1-liner for day 1? 👀

[mossy trout]

😭😭😭

[hollow summit]

wait until this guy find out about part 2 😭💀❌❌

[mossy trout]

My day 2 is bugging

[agile orbit]

Ahh, no biggie - just gonna use a dict for conversions

[hollow summit]

nuh uh

eightwone is expected to return 81

[woeful topaz]

i somehow managed to solve this in 32 lines 💀

now i wonder, what is the shortest possible way to get the solution without using semicolons

[modern terrace]

yes

it works with the general .replace() approach

[agile orbit]

Yeah..?

[modern terrace]

a more not great test case would be 5twone which should return 51

[agile orbit]

Just need to ||replace all instances of words -> numbers, you can use a dictonary for the mappings||

[modern terrace]

good luck with that approach

[agile orbit]

Thanks!

I'll try my best

[modern terrace]

||5twone should equal 521/5tw1, not 52ne||
||7oneight should equal 718/7on8, not 71ight||

[agile orbit]

Yup, I read the examples given 🙂

[mossy trout]

can someone explain why my code is working but failing

it works on the base case

it gets me the right games

and adds their IDs

yet its wrong

[modern terrace]

you shouldn't add the numbers together

that's the problem in your code

[mossy trout]

Seriously?

Determine which games would have been possible if the bag had been loaded with only 12 red cubes, 13 green cubes, and 14 blue cubes. What is the sum of the IDs of those games?

[modern terrace]

sum of IDs is okay

sum of reds, no

sum of greens, no

sum of blues, no

they're each separately compared to the 12/13/14 count

[mossy trout]

ok

[hollow summit]

if r>12 or g>13 or b>14 then it's invalid

[modern terrace]

Game 101: 6 red, 7 blue; 5 green, 9 blue, 7 red; 9 green, 2 red
``` this should be a valid game

[mossy trout]

OHHHHHHH

i see

so each ;

is a different iteration ehck

[lusty rover]

See #1180008645384216606 for the golfing

[agile orbit]

Thank you!

[magic oxide]

oi

one liners aren't necessarily golfing

[upper pike]

Can I get advent of code for actual morons?

[lusty rover]

🤨

[magic oxide]

my one liner passed fiddle code review

well i'm counting one liner as after the file.read().splitlines() in the parser

[covert arrow]

oneline golfing is pretty cool though

[magic oxide]

i'm meh on it

i like regular golfing, and i like eval golfing (has to be an expression) but for some reason just that not interested in one line golfs

[hardy zenith]

I have two of these and they've been fantastic

[sinful pewter]

Hey guys, I need some push to get started. Fear of failure is keeping me back

It always happen when I start something from scratch

[unique osprey]

Fear of failure for AoC? Failure to do what though, get on the leaderboard? Solve the puzzles?

[sinful pewter]

Everything, literally everything but more so solving the puzzles themselves.

[scarlet wasp]

Just a matter of perspective. If your focus is to learn things (and maybe even have fun 🙂 ) rather than "winning", then you basically can't fail. I've been coding for 10+ years and still managed to learn something day 1 😄 that's a win in my book. Even though I got a horrible time. But yeah, use the community, ask stuff, keep learning

[unique osprey]

Yeah really there's nothing to lose by doing the puzzles.

And hey, the leaderboard top 10 is pretty darn hard to get on to

[hidden musk]

by the time you're good enough to get on the leaderboard you've definitely gotten rolled enough times anyway

[sinful pewter]

I will get started with day 1, I will report it back here. I wanna have fun. I love the fact that strangers from all around the world are just focused on this one event

[hidden musk]

strangers around the world that are focused on this event are focused on this event

[sinful pewter]

🙂

[scarlet wasp]

Don't be afraid to ask for help if you get stuck. All about learning

[sinful pewter]

Got it will do

Thank you

[agile orbit]

@modern terrace #1180009538544488478 message
Finally did it, you were right in saying about those edge like cases!

Appreicate the help

I tottaly forgot about regex....

[compact wasp]

eh

1 year of getting rolled isn't that much

though i'll probably fall off the board once i can't yolo things

[magic oxide]

you say this every time and then still leaderboard

[compact wasp]

i only leaderboarded like 4 times last year

and those were not even high leaderboard

[magic oxide]

you weren't there at the start everyday last year

[compact wasp]

true

[hollow wharf]

talking about day1 p2?

[woeful topaz]

about today

[hollow wharf]

ah

[woeful topaz]

but i feel humbled by someone with about 10 lines of code

[hollow wharf]

i find part 1 harder than part 2 haha

well this is my solution for today ||https://github.com/DownDev/advent-of-code/blob/main/2023/02-2.py||

[sour trout]

Does anybody know how would i add my repo here? https://github.com/Bogdanp/awesome-advent-of-code/

It seems a little bit confusing to me as never contributed to something

[agile orbit]

https://github.com/Bogdanp/awesome-advent-of-code/pulls
open a pull request

[sour trout]

lol, for me in day 1 part 2 it was a little bit weird because of the: "twone, eightwo"

[sour trout]

I did

I dont know if i did everything correct to be honest

It got added YES

All fine thanks

[hardy zenith]

wow today was way easier than yesterday

yeesh. I'm way behind on my private leaderboards because of timezones

[halcyon pulsar]

i would be curious what timezones the top people are on

whats an ideal timezone?

[compact wasp]

central time to pacific time

actually you could probably go west to idl and still have a good timezone

[proven quiver]

Is this a place to ask for help with AoC/

[sour trout]

I could literally solve the day 2 - part 1 without coding but it would take a lot of time

It was easy

[proven quiver]

Here is my attempt for day 1 part 2 but it's too low. Anyone know where I've gone wrong?
||https://hastebin.com/share/revevoxuzo.python||

[magic oxide]

||if you break after checking if one is before the first digit, then it fails if there's like an eight before that, so for example on the input 'eightone1', ||

||also you're off by one for enumerating nums since enumerate starts at index 0||

if you wanna discuss further i'd recommend moving to #aoc-solution-hints

[halcyon pulsar]

||also no 8||

[ivory dagger]

For day 2 part 2, you should only calculate for the games you kept after part 1 or for the entire input?

[low condor]

the example input makes this clear :)

[ivory dagger]

it does! thank you!

[low condor]

Advent of Code is partly Advent of Reading Comprehension... until even reading doesn't help lol ||cough cough day 1||

that's a story right there

[magic oxide]

the new type of comprehension

we have list comps, set comps, dict comps, gen exprs, and now read comps

[compact wasp]

someone pinging the mods to say the n word 💀

just why

[frigid mural]

Too much time on their hands and not enough brain cells

[compact wasp]

real

brick for a brain

[magic oxide]

the n word thing or aoc?

[compact wasp]

n word thing obviously

[magic oxide]

if you're a beginner i'd say start with ``print('Hello, World!')` first

on a serious note, if you do mean aoc, i'd say give it a try, and if you can't then come back to it later/look at other people's solutions for some insight

tho this year's day 1 is a bit hostile so maybe skip to day 2

[compact wasp]

nah that's mid

do exec(bytes(ord(c)%i+32for c in'񳪪򛳴񵑦򅔢󖅞񶊂񗋯򍄦'for i in b'efg'))

[magic oxide]

is there no way to get rid of that +32 by using higher ord bytes and different unicode values

[wicked elm]

so for some reason I had a rule that both parts of my solution would be standalone, and I've already broken it in day2

[terse comet]

can we not go full esoteric while beginners are watching

[magic oxide]

fair enough

[alpine tree]

hi

[untold briar]

Hello, is 54431 close to the result of challenge 1?

[alpine tree]

how do i sign up for aoc?

[untold briar]

Go to aoc-faq channel just below

[peak dock]

different people get different inputs

[wicked elm]

did you find that somewhere or do people really make these themsevles

[untold briar]

This challenge shouldn't have been that hard... I'm scratching my head here

I tried checking them one by one to see where the error was but my eyes gave out after the 50th

[compact wasp]

[untold briar]

And there are 950 more to check 😂

[alpine tree]

hm okay so how does this work, im on the aoc website i clicked on the first puzzle, what do i do?

[untold briar]

ff6dhvzmdrgt should be 66 right?

[untold briar]

You must login using one of the authentication methods

Then read through the puzzles and solve them

[low condor]

||yup||

[alpine tree]

woah the puzzles are more of an english challenge than a programming one

[compact wasp]

real

[untold briar]

Could I post a file containing my input and the number I found here and ask someone who is willing to check where my mistake was? Checking 1000 entries one by one manually is just... my eyes are spinning haha...
Format will be:
entry1:num1
entry2:num2
For example:
jninepzpgtzq7four5:95
fourvjjrttlvdtfour8qxdvlg22two:42

[alpine tree]

how do y'all structure your environment for the aoc puzzles?

[hollow wharf]

just input.txt file and python file

[compact wasp]

poetry for dependency management, have my own scripts for pulling inputs, challenge texts, and examples at midnight

automated submission as well

then shortly before the thing drops i copy a template

[alpine tree]

so do you make a folder like that for every day?

[low condor]

How about if you just post your code and I check it against my input? Then I can give you either a hint, or the line that was bad

[hollow wharf]

not really

[low condor]

Probably don’t post your code here though - do it in #aoc-solution-hints

[alpine tree]

so how u do it

[hollow wharf]

im working in the same folder each day

[runic isle]

Woah

[alpine tree]

so u dont even save the previous solutions?

[hollow wharf]

i do, i also save them to github

[compact wasp]

a few seconds matter a lot for the first few days

[alpine tree]

so you make a repo for each day?

[hollow wharf]

no, its one repo

||https://github.com/DownDev/advent-of-code||

[compact wasp]

he just uses commit history to see previous code fr

[alpine tree]

so how do i automate getting the input?

[hollow wharf]

imagine

[hollow wharf]

well i saw people use some libraries

!pip aoc-lube

[marsh current]

aoc-lube v1.0.0

Proper Advent of Code lubricant.

[earnest forge]

!pip aoc-helper

[marsh current]

aoc-helper v1.8.12

A helper package for Advent of Code

[earnest forge]

:P

[rapid wolf]

I just copy and paste

Got notepad open

[hollow wharf]

same 😊

[earnest forge]

!pip aocd is a thing too

[marsh current]

aocd v0.1

dummy name to pull advent-of-code-data

[compact wasp]

requests.get(f"https://adventofcode.com/{YEAR}/day/{DAY}/input", cookies={"session": AOC_SESSION})

[sour trout]

What does exacltly go in : "AOC_SESSION"

[compact wasp]

your aoc session token

[magic oxide]

you grab it with inspect element

salt die's repo has a neat graphic

[woven oriole]

beep bop (session cookie)

[earnest forge]

Yes! 🙃

[hallow birch]

Thanks!

[sleek remnant]

is it normal that the page looks like this? I've seen it being an actual christmas tree somewhere

[untold briar]

Hmm, do you guys also think that day2 is easier than day 1?

[earnest forge]

Yes

[untold briar]

Day1 part 2 was a bit janky, I had to debug my code a lot until it worked

Day 2 I blew right through without even debugging lol

[earnest forge]

Calendar art is different each year and is unlocked by solving puzzles

[naive stone]

the picture fills in more each day

[untold briar]

Yeah if you solve day 2, it will change a bit

[earnest forge]

Tree was 2015 btw

[untold briar]

Oh day 1 was a trebuchet lol

I only noticed it just now

And day 2 is a floating island I guess?

[outer chasm]

most definitely

[sharp tapir]

idk

depends

managed to do both

needed some help figuring out how to parse a string

but i had the logic on lock

[noble skiff]

day 1b had the edge case with overlapping words

day 2 doesn't really seem to have any edge cases that caught a lot of people

[cunning scarab]

Not really, parse part 1 right and you got part 2

[noble skiff]

Can't tell if you're contradicting me or agreeing with me 😛

[earnest forge]

Looks like they're concurring

[cunning scarab]

Agreeing lol

Not really refering to "any edge cases that caught a lot of people"

[glad tartan]

There is a bug in today's page:

<li>Game 2 could have been played with a minimum of 1 red, 3 green, and 4 blue cubes.<li>

[torpid karma]

This is right?

[terse comet]

what exactly is the bug?

[glad tartan]

unmatched <li> tag

[earnest forge]

(that's actually just how HTML works)

Oh wait yeah no it creates an empty li lmao

I mean I'm pretty sure browsers will just ignore it Nope it's just that game 3 renders over it lmao

[terse comet]

oh lol

eh, if it works it works

[earnest forge]

Damnit Eric

(BS4 doesn't understand it lmao)

can someone go pester him to fix it

[terse comet]

why are you trying to bs4 it in first place xD

[earnest forge]

Someone asked for 'read the puzzle in the console' as part of my library so I'm adding that

I thought I managed earlier but I forgot lists lol

[oak root]

Hey, I need some help with a relatively simple python program and im willing to pay anyone who can make it right now, dm me if youre interested

[royal gull]

!rule 9 6 no paid, but you can get free help in #python-discussion

[marsh current]

Rules

6. Do not post unapproved advertising.

9. Do not offer or ask for paid work of any kind.

[minor ivy]

For part 1 of day 2 is the format it's using sets or lists??

I'm confused

because if it's sets then indexing won't work

[terse comet]

neither?
the "format" it gives you is just plain text, you have to parse it yourself, and it's up to you to determine whenever you'll want to parse it as tuples, lists, sets, dictionaries etc.

[minor ivy]

Ohhhh

alright

[earnest forge]

@hallow birch I fixed the console reader to actually render lists now, you should update if you want to use that

[livid gull]

If anyone wants to see where they are globally if you've gotten points for being in the top 100 but arent in the main 100 overall here you go.
https://github.com/SaiMonYo/Global-Leaderboard-AOC

[low condor]

You can show multiple rules at once through the bot ? That’s pretty cool

[sleek cave]

i finally understood aoc 21 day 22 (reactor reboot)

coding it in python tomorrow

my redemption after 2 years

[low condor]

This is my nightmare day

Easily the worst day for me of recent Advent of Code

I hate 3D puzzles as I can never visualise them, but when paired with a solution that passes example but not real input, and numbers so large that debugging is nigh impossible… Yup, I failed this one HARD

[frigid mural]

Did a little catchup for Rust but now we're all good

[worthy horizon]

Grrr. Part 2 of day 2 might not be possible with spreadsheets
Maybe I'll try to find a way to find a maximum with regex

Anything repeating is generally a problem

[zinc kindle]

I'm trying to complete old puzzles and I'm failing 2019 day 22 part 1. My code works for the sample inputs but not the actual input, and I can't find hints online (only solutions). Can someone help me with this?

[terse comet]

does it needs to be reproducible and work for any input, or can you do some manual steps?

[worthy horizon]

Manual steps are fine I think. But I'd rather keep them to a minimum

[terse comet]

you can probably solve it using a pivot table

[zinc kindle]

Never mind - I misread it, thought it was asking for cards[2019], was asking for cards.index(2019). All that time trying to figure out why my algorithm was wrong.

[somber trench]

How long will this be open for after the 25th?

[hidden musk]

define "this"

[somber trench]

The advent of code days, the ability to do them?

[zinc kindle]

Until the owner dies and stops being able to pay for the server? You can still access the oldest ones in the archives.

[hidden musk]

forever, maybe

[somber trench]

You can? Dang thought it was unavailable after the 25th coolio

What year was the first one done?

[magic oxide]

even after the sun dies?

[hidden musk]

maybe

[zinc kindle]

https://adventofcode.com/2015/events
2015

[magic oxide]

damn

[somber trench]

Almost 10 years

[terse comet]

"only" 8 years

[somber trench]

Next year will be 9

[zinc kindle]

If you ocpmlete a problem over 24 hours after it comes out, your time is just listed as >24h instead of your actual time. Aside from that, you can complete them whenever.

[somber trench]

Cool cool

Day 2 is a pain idek where to start

[woven sable]

im trying to do aoc in kotlin

i was stuck on day 1 part two for a good 30 minutes 😭

[cunning scarab]

"eightwo"

[woven sable]

oh my god

[royal gull]

Yah. That nearly ended me.

[woven sable]

for me it was j coding the ||finding the digits as words from the start and end when they were combined bc i tried replacing the words w their digits||

i had to use regex and im def not proud of my solution

[cunning scarab]

Yeah I did that too.

[covert arrow]

I want to do one in piet but there is no way I'm doing day 1 part 2

[cunning scarab]

But I'm glad everyone else struggled as well I nearly had a stroke.

[woven sable]

i think my worst day in 2021 was the beacon scanner

[cunning scarab]

Was that with the grid?

[woven sable]

i'm pretty sure that day took me 6 hours

yeah i think so

the one w the different 3D scanners

[minor cave]

tempted to otn oneightwo

[hidden musk]

twone

[cunning scarab]

twoneight

[lapis pilot]

This is my first year… how much harder does this get???

[cunning scarab]

uhm get ready for 2d array's around day 5

[lusty rover]

[woven sable]

the first day is usually not this hard honestly

[cunning scarab]

I really like the rocket fuel one

[minor cave]

!otn a oneightwo

[marsh current]

:ok_hand: Added oneightwo to the names list.

[lusty rover]

A lot
Check out last year’s last problem: https://adventofcode.com/2022/day/25

[lusty rover]

Thank

[woven sable]

i think it peaks around day 18-20?

usually they don't make it as hard on day 25 bcs it's christmas

[hexed axle]

I should give part two of day 1 a shot, it looks like everyone had ™️*f*u*n*™️ with it

[woven sable]

idk how long i'll be able to do this in kotlin

before i switch back to python

[lusty rover]

Why did you skip it?

[royal gull]

It actually was quite fun, just a good ‘need to think about what’s going wrong’ type problem

[lusty rover]

(How well do you know Kotlin?)

[woven sable]

i can code most things in it

[hexed axle]

I did a ctrl+f to see if my hunch was right then I called it good. I didn't have time and I have an old textbook that I want to use on it

[lapis pilot]

Well fuck

[woven sable]

like i can code everything if i search some things up

i'd say im pretty comfortable with it

[lusty rover]

I actually accidentally completely skipped all of the edge cases with my original implementation

[woven sable]

im gonna attempt day 2 rn

[lusty rover]

Good luck!

[woven sable]

i would've been so happy to do day 2 in python

looks like u could use ||match/case|| so the it

with*

[lusty rover]

(No one is stopping you)

[woven sable]

ty!

true

[lusty rover]

I did

Let me know you wanna see

[woven sable]

wait can u send ur solution im curious how u did it in python

yeah fs

[lapis pilot]

What are you using @woven sable?

[molten fulcrum]

For day 1 puzzle 2, my code works perfectly for the examples, but it says I'm getting a too low value for the real data. Anyone knows about some common issues people run into on that question?

[lusty rover]

Jinx

[lusty rover]

https://github.com/shenanigansd/scratchpad/blob/main/events/advent_of_code/2023/01/python/aoc_2023_01.py

[covert arrow]

||oneight should be 18||

[woven sable]

kotlin

[woven sable]

ty!

what does math.prod do?

i might try day 2 using ||classes||

[molten fulcrum]

It seems to be parsing that correctly too

[covert arrow]

I don't think there were any other edge cases

[lusty rover]

Returns the product of multiplying every number in the list that you give it

[molten fulcrum]

Hmmm, strange. Maybe it's not a logic bug then, but just me being horrible at writing groovy code

[cunning scarab]

Should be a build-in feature

[woven sable]

ohh

[lusty rover]

It can also be ||frontloaded||
i.e., ||twone is 21||

[lusty rover]

It… is

!docs math.prod

[marsh current]

math.prod

math.prod(iterable, *, start=1)```
Calculate the product of all the elements in the input *iterable*. The default *start* value for the product is `1`.

When the iterable is empty, return the start value. This function is intended specifically for use with numeric values and may reject non-numeric types.

New in version 3.8.

[covert arrow]

Stdlib is close enough tbh

[woven sable]

also ||eightwothree should be 8wo3 or 823 however u did it|| but i'm sure u alr know that

[molten fulcrum]

Yep yep, that's working like it should. ||"twone" gives 21||

[cunning scarab]

Yeah I meant not stdlib

[cunning scarab]

Yeah fair ig

[covert arrow]

It's pretty niche so adding it as a builtin would just be bloat

[somber trench]

I've given up on this year so imma start at 2015 and move up lol

[cunning scarab]

But sum is the same

[lusty rover]

How far did you get?

[cunning scarab]

Guessing d1p2 lell

If you've done d1p1 then you can do d2p1&2

[covert arrow]

Sum gets a lot more use than prod

[cunning scarab]

[somber trench]

On this year or 2015?

[lusty rover]

This year
Was just curious how far you got before you gave up

[somber trench]

Day 2 part 1, Day 3 ain't out yet so

[lapis pilot]

Kotlin… my daily new language lol

[lusty rover]

Oh
Okay, you’re getting it

[lusty rover]

How’re you liking it?

[somber trench]

Getting what?

[tepid cairn]

I wanted to do this year in Agda but Agda doesn’t have any decent IDE tooling 😩

[lapis pilot]

Oh.. I meant my daily language that I’m hearing of for the first time

🫠

[lusty rover]

So?????
REAL programmers use ed

Who needs to see their file while they’re editing it????

[cunning scarab]

My daily language is dutch.

I don't recommend it

[lusty rover]

[lusty rover]

Ah

At work I’m rebuilding a Spring app and I’m rewriting the Java to Kotlin

It’s been nice

[lusty rover]

Getting the problems done
Getting things solved
When you said you gave up I thought you were still on like d1p1
But you’ve got almost everything so far solved

[somber trench]

Idk where to start for day 2 that's the issue, ik what to do for it idk how to implement it is the thing, so ik the theory, not the practice ig

[lapis pilot]

Wait… are we not supposed to research methods on Google?

[somber trench]

You just need to get the answer with any means other than chatgpt or other ai I think

[lapis pilot]

Phew 😮‍💨

[magic oxide]

you can google as much as you want

[cunning scarab]

You're free to use it after the leadboard has been filled up.

[molten fulcrum]

I rewrote the logic to be less messy, and now it works

Time for some OCaml

[somber trench]

Ah, mk

[sleek cave]

this was the day i gave up on aoc 21 lol

3D is super hard, but for me it's also incredibly satisfying

esp after having learned some basic CAD for an intro tech design class

i read up on quite the elegant solution for day 22

i was overthinking this whole time

no splitting cuboids needed :D

[woven sable]

day 2 was sm fun w kotlin

i got lucky bc the approach i took for part one was ||finding the max amount of cubes drawn frok all the turns w each color|| so part two was easy

[woven sable]

oh it's not my daily

i use it for robotics mainly

my daily is fs python i love python

j exploring new languages w aoc since i've already gotten all 50 stars once

[lapis pilot]

Same, same

[lusty rover]

What in the name of….
how many hecking repos do you have??

[lapis pilot]

I’m curious to look at the other languages

[jovial jay]

i'm challenging myself to use a non-generic language to do the AoC

[tepid cairn]

it put emacs shortcuts in my vscode 😩

[tepid cairn]

have you looked at the daily language roulette

[jovial jay]

no

[hexed axle]

AoC in PowerPoint?

[lusty rover]

at least it wasn’t vim

[tepid cairn]

it’s a random language from a pool for every day you’re challenged to do the puzzle in 👀 yesterday it was groovy, today it’s ocaml

[woven sable]

where's the roulette

[jovial jay]

I'm doing it in G'MIC. i posted my solution for part one of day 2, but as soon as i see part 2, i decided to rewrite the solution.

[woven sable]

wdym by a non generic language

[hexed axle]

It's announced each day, and pinned both here and in the spoiler thread

[tepid cairn]

it’s pinned in the thread every day

[jovial jay]

see above. i'm doing it in G'MIC. a language specifically designed for creating/editing raster graphics images.

[lusty rover]

Announced in this channel like 10 minutes before the timer unlocks

[hexed axle]

That is quite a challenge, nice!

[cunning scarab]

Day 1 is missing the pin

[tepid cairn]

ah, I see

[woven sable]

is advent of code at 12 am est

im jealous of ppl on the west coast

[lapis pilot]

I think it’s funny that Ocaml describes itself as “industrial-strength”

[hidden musk]

it's used by like 2 companies at least

[lusty rover]

Yes

[lusty rover]

Which two?

[hidden musk]

semgrep, jane street

[lusty rover]

Anyone important?

[hidden musk]

me

[lost nest]

guise

did anyone unlocked problem 3?

[peak dock]

&aoc c

[shut trellis]

Day 3 starts <t:1701579600:R>.

[leaden tartan]

Got in right under the wire!!

[celest rain]

i learn a new language every time i do aoc, shit's crazy

[compact wasp]

the beta testers did

[magic oxide]

looks like I can't make it today either

[visual nebula]

[harsh heart]

oh shit

advent of code started

[leaden tartan]

its currenty 4:30 AM. Do I push through and try to solve day 3 before bed or sleep and work on it tomorrow

well

today

[compact wasp]

just sleep

[earnest forge]

Nearly missed waking up for today lmao, luckily I'm up now

[minor cave]

Language Roulette: Day 3
The language is ... Haskell

[placid lake]

some people are going to have a field day today

[lusty rover]

oh no

[peak dock]

my time has come

[compact wasp]

😈

[lusty rover]

I tried Haskell last year

I couldn't read the file

[compact wasp]

haskell is fun

[sharp panther]

deciding whether to do this problem now is going to depend on how hard the parsing is :(

[odd wigeon]

https://tenor.com/view/dancing-coffin-dancing-pallbearers-funeral-dance-gif-16837090

[compact wasp]

skill issue tbh

[peak dock]

but theres a readFile function

[sharp panther]

10 seconds...

[lusty rover]

[shut trellis]

<@&518565788744024082> Good morning! Day 3 is ready to be attempted. View it online now at https://adventofcode.com/2023/day/3. Good luck!

[white laurel]

https://tenor.com/view/the-office-steve-carell-michael-scott-omg-oh-my-god-gif-4767816

[sharp panther]

...yep we have some parsing work, this is a tomorrow me thing

[valid brook]

ayy

[minor ivy]

Day 2 is hard as fuck

I can't do it

[potent pumice]

i'm a bit surprised language roulette isn't automated lol

[lime steeple]

yikes

[lusty rover]

sheeeet

I don't even know how to do this in Python

[sleek cave]

tf is with aoc and parsing this year

[shadow musk]

good morning 😭

[white laurel]

ok this one is need time XD

[lusty rover]

The lack of a thread going up is making me worried about part 2

[lime steeple]

we all read the problem and decided to go to bed

[sharp panther]

the lb is full :)

took surprisingly long

[hexed axle]

11.5 minutes, on day 3

[lusty rover]

SUPRISINGLY?!

Have you SEEN the problem

I'm stuttering over here

[sharp panther]

surprisingly, as in for a rather early day

the problem is not easy, yes

[compact wasp]

fuck i was scrolling reddit and didn't see the time

i should've seen the timestamp on my discord message

[fleet fox]

day 3 is... not ok

[compact wasp]

doesn't seem that hard tbh

[fleet fox]

its making my brain rot

[compact wasp]

though i get the feeling that these are intentionally designed to be difficult for gpt to do

[covert arrow]

i was watching owc 😔

[earnest forge]

I'm dumb

[white laurel]

can't figure it out

[earnest forge]

I spent like 10 minutes on what turned out to be a name shadowing error

[woven solar]

in the middle of solving day2 I realised I have an unnecessary third nested loop in my day1 solution (that just didn't affect the day1 result). have to be really sure these numbers are adjacent, y'know?

[earnest forge]

Lmao

[hazy ermine]

man imagine the amount of effort required to create these puzzles

[hollow wharf]

I'm going to bed 💀 😭

This is a problem for future me

[iron olive]

I dozed off; how is day three? hard?

[hazy ermine]

a little, yes

[iron olive]

Harder than day two?

I mean, that's to be expected

But how much?

[hazy ermine]

yea its more annoying to process the input than hard

[iron olive]

Depending, I'll go straight to bed and leave that for tomorrow me

Well, that's for tomorrow me to solve

Nighty night

[woven solar]

about the same as day2, IMO, and way easier than day1 🥴

[candid berry]

[hazy ermine]

why did they mess up the difficulty lol

[iron olive]

I'm tempted to peek at it

[candid berry]

day 1 just needed better examples

[iron olive]

But if I do, I'm solving it

[candid berry]

it really wouldn't have been bad if the examples had properly covered the edge cases

[hazy ermine]

i used another method (without any replacement) and i didnt face any issues because of the overlapping text, so ig its easy/hard depending on the method

[woven solar]

yeah, that's fair. technically the day1 example does have the problematic case, but it doesn't spell it out in the text, so you have to notice it (which is a hard thing to do when skimming the text as fast as possible)

[candid berry]

it points out what the problematic case is, but it doesn't tell you how to handle it

[woven solar]

yeah, that's true

[candid berry]

there isn't an example where ||the rightmost digit is part of an overlap||

[earnest forge]

(in fact, it's never used at all)

[candid berry]

I dunno. I didn't find 1 hard, just underspecified

and that's frustrating

[earnest forge]

The thing I disliked about 1 was that ||you had to change your parse for part 2, rather than your logic||

That's less fun IMO

[hazy ermine]

well tbf the ||parsing is the logic here||

[earnest forge]

Ig

[compact wasp]

i should really take the time to add stuff to my template

like a grid would've been really nice to have

with stuff for finding neighbors

[deep ferry]

bruh i had a dumb bug for problem 1 today

because

||file.readlines() includes the \n after every line||

[feral hazel]

yeah i've written for dx, dy in product(range(-1, 2), repeat=2) and for dx, dy in (-1, -1), (-1, 0), (-1, 1), (0, -1), (0, 1), (1, -1), (1, 0), (1, 1) far too many times lol

[deep ferry]

so like file.readlines() != file.read().splitlines()

i'm never using file.readlines() ever again

[deep ferry]

i use 2 arrays

and a for loop for it

imo it's cleaner

[livid gull]

time to go back to sleep

[feral hazel]

ah i prefer the product
when i'm writing it i'm usually already like 4 levels of indentation deep and every bit helps lol

then again last year itertools became the hammer i've struck far too many nails with

[woven solar]

i never use it, yeah. it gives raw lines, with the newlines at the end.

[stable shale]

Took me a bit to realize ||duplicate entries of the same number is allowed with different symbols||
That aside, not too bad

I really need to make a commons library though, the number of times I've had to type out an adjacent coordinates list

[brazen bridge]

how can I see what number I finished as?

[woven solar]

leaderboard-personal times

[velvet stag]

that's cursed is it true?

does it like happen once, so it can't be spotted easily?

[earnest forge]

...why is this a surprise to anyone lol

[stable shale]

¯_(ツ)_/¯

[earnest forge]

Isn't that just the default interpretation of the problem statement for p2

[woven solar]

yeah, not sure what you mean, it seems normal to me that a number can ||be adjacent to multiple gears, contributing to all of them||

[stable shale]

For P1

[earnest forge]

For P1, ||no you don't count the number twice||

[stable shale]

Looking back, yeah, definitely. Probably read the problem badly

[earnest forge]

(also ||P1 doesn't mention gears||)

[velvet stag]

it's a surprise because i'm looking at my input and i don't see an example

[stable shale]

No idea how frequent it is, but didn't work for my actual input, reread the question, took a gamble and ||reset duplicate tracking after each symbol instead of having it cumulative||, and it worked, so no complaints here

[woven solar]

that might be true actually, I don't see an example of a ||number with >1 adjacent gear|| in my input either. weird; maybe they throw a bone to people who assume there's a strict 1-2 relationship

[stable shale]

I'm glad that happened to me on P1 rather than P2, because I'm sure I would've spent a lot longer debugging

[compact wasp]

i should create a 2d regex tool

[stable shale]

YES!

[compact wasp]

would get #1 on these kind of problems fr

[woven solar]

this day is kind of a case of "i feel like I should write an utility to make days like this easier, but the problem is weird enough that it doesn't obviously generalize". like, what would I implement, ||neighbour-finding for a set of rectangular cells||?

[compact wasp]

literally sub 1 minute times and get people to think i'm cheating

[odd wigeon]

80 more lines, another challenge done

[compact wasp]

finding neighbors is used a bit in aoc tbh

[woven sable]

||could a possible edge case be two numbers that are the exact same which are adjacent to a symbol? ||

[stable shale]

Accused of AI immediately

[compact wasp]

fr

but it's just me using 2d regex

[woven sable]

or is that not sum to worry about

[odd wigeon]

I haven't seen a situation like that in my input that would effect the result.

[woven sable]

like ||35*35||

alr awesome

cuz idw recode anything

[earnest forge]

Istg 2D regex would be like what people think regex is like lmao

[odd wigeon]

as long as you're not only doing unique values

[compact wasp]

fr

i remember i thought of the idea when i made boxscript for one of the code jams here

[stable shale]

Wait....if we have 1D regex and we can make a 2D regex crossword with it. 2D regex means we can make 3D regex crossword!

[compact wasp]

but decided to just manually code the 2d syntax

[lost nest]

i have a question here

are numbers counted as symbols?

[woven solar]

you can make a 3d regex crossword with 1d regex too, I don't see why not

but sure, it'd allow for more cursed stuff

[earnest forge]

(what's a regex crossword?)

[lost nest]

?

[empty eagle]

wait do i have to || count the number twice if its adjacent to 2 symbols for p1||?

[stable shale]

Crossword, but with regex across/down

[stable shale]

Yes, at least I needed to for my input

[empty eagle]

smh

ok

[earnest forge]

No, at least not for my input

[earnest forge]

That sounds horrifying. Do you have any examples?

[woven solar]

see https://regexcrossword.com/ (although those dumbasses recently reworked their decent UI into some sort of webapp abomination; who thought that's a good idea?..)

[stable shale]

^ that

[earnest forge]

Lmao

[stable shale]

If I remember when I wake up tomorrow, I'm gonna modify my solution to find, and tell me whenever that ||number with multiple gears adjacent|| occurs, I really wanna know how frequent it is

[woven solar]

never for my input

[earnest forge]

Never for my input

If you wanna send me your input I can find out for yours

[woven sable]

my test data works but my input data doesnt work 😭

[earnest forge]

Rip

[humble escarp]

2nd day in a row that I start without regex, delete everything, use regex.

[hazy ermine]

lol

[humble escarp]

the dumbest part is that I was ready for it: AoC on left screen, editor in central screen and Regex101 on right screen. Well, it won't happen 3 days in a row.

[hazy ermine]

ig its easier without regex here

[stable shale]

https://paste.pythondiscord.com/raw/MJ7A

[compact wasp]

until i make 2d regex

[woven solar]

he says the gondola lift will take you up to the water source
come back tomorrow for AoC 2022 day 16 2.0: electric boogaloo 😩

[humble escarp]

My regex for today || /(?<num>\d+)|(?<symbol>[^\.\d\s]+)/g ||

[stable shale]

My regex setup was real dumb for today, but hey, it worked

[compact wasp]

once i make 2d regex it'll be over

[hazy ermine]

2D regex wtf

u can do that?

[compact wasp]

i don't know yet

[stable shale]

Not yet

[compact wasp]

but it'd be a sick project

[stable shale]

But if we can, oh the possibilities

[earnest forge]

This is pretty cool but I hate that I have to drag letters into place rather than just typing like a normal crossword

[stable shale]

(and sanity saving)

[compact wasp]

2d regex would trivialize this lol

[stable shale]

Yeah, that's the UI update

Look for any challenges category with a keyboard icon

Those use KB instead of the drag & drop

[earnest forge]

Never for your input either

[stable shale]

HMMMM

¯_(ツ)_/¯

[hazy ermine]

so 2 gears are never adjacent to the same number?

[earnest forge]

(also technically I'm checking something more general than what you said - I'm checking that ||no two symbols share a number|| rather than just checking gears)

[stable shale]

I have no idea, and to protect my sanity, I'm not digging further

[woven solar]

it's possible to enable keyboard, which I know because it's obligatory for some player puzzles (why isn't it default?..)

[stable shale]

Yeah, I meant ||symbol||, not ||gear||

[earnest forge]

How do I enable the keyboard lol

[stable shale]

Who knows

[earnest forge]

Oh ok once you play one that requires keyboard you can just use keyboard lol

[stable shale]

Yeah

[earnest forge]

Though you can't use arrow keys -_-

[stable shale]

The ones with KB are indicated with a KB icon, and iirc the player made ones are KB (or can be made KB by the creator?)

[stable shale]

Tab & Enter?

The old UI was much better

KB always, arrow keys worked

[earnest forge]

Wait enter

Ugh

(no, enter does nothing)

[stable shale]

Yeah, I dunno then, gone?

Wish they'd add a old.regexcrossword.com lmao

[woven sable]

my solution didn't work

bc for some reason i didn't calculate the last row

fun challenge

rlly proud of how i did it in kotlin

[compact wasp]

the hexagonal ones are kinda nifty

[vestal breach]

advent of regex

[hazy ermine]

how are people using regex for day3 💀

but yes, regex was involved extensively even in the previous years i believe

[naive stone]

regex was very handy for today. Just solved part 1 finally

part 2 seems tricky

[woven solar]

i used it ||to quicky grab all numbers in a line along with their spans||

[naive stone]

i'd have had to write a lot more logic if not for ||re.finditer||

[hazy ermine]

ahh

[placid lake]

huhh i did not think of using regex

[woven sable]

||same here, and then i checked the adjacent y and x values by if they were in the span of the numbers||

[placid lake]

i went ||numpy||

[woven sable]

regex saved me fs i was lost asf

for a good 30 min

[naive stone]

fancy

[woven sable]

i tried ||searching for symbols thru numbers|| first

[hazy ermine]

ok this is a change in perspective

[woven sable]

how'd you do it

[young tree]

Numpy has a builtin for finding neighbors?

[woven solar]

@signal copper slides into chat: let me tell you about our lord and saviour, scipy.ndimage.convolve

[woven sable]

part 2 was either rlly simple or rlly tricky

[woven solar]

(i didn't actually use numpy this time, though)

[woven sable]

if u ||searched for numbers and then found a symbol and marked it down, it's def harder||

[placid lake]

nah, was useful in other ways

this would've been fairly easy ||as a graph problem using networkx|| too i think

would be pretty overkill though

[woven sable]

damn yall r crazy smart

i can't imagine doing these problems in ways like some of u do

w google sheets and sql and all

[placid lake]

will send solution after a little bit of cleanup

[signal copper]

the naive solution is also fine here, the inputs are very small

||iterate through the rows, find numbers, check adjacent cells||

[hazy ermine]

add spoilers please

[woven sable]

what was the more efficient way?

[young tree]

I dont understand graphs very well but would u like || pair each star with its neighbouring ints and proceed from there?||

[woven sable]

i did that but ||found symbols then looked around for numbers||

[signal copper]

||there are fewer symbols than numbers, but maybe parsing the numbers is more difficult if you find the symbols first, e.g. having to look both left and right of a digit instead of only in one direction||

[woven sable]

yeah ||what i did was first parse thru and find all the numbers w their span using regex, then iterated again until i found a symbol and looked at all the adjacent cells and see if they're within a span of a number||

[signal copper]

||if you look at the indices of where the numbers are, this immediately tells you which cells to check for symbols without needing to iterate again through the input||

[woven sable]

that's true

i should've done that

[pine ravine]

is there some weird interpretation? my friend and I both pass on the test data but not on the main data

[woven solar]

aoc chat be like: ||[REDACTED] [REDACTED] Class: Keter [REDACTED] 269[REDACTED] civilians were executed [DATA EXPUNGED]||

[pine ravine]

at this point i dont even know how to debug, my code gives exactly what i would expect

[placid lake]

||i had an issue with numbers on the right edge||

[vestal breach]

236.236
...*...

[pine ravine]

yap that would be it

[vestal breach]

👍

[pine ravine]

wait no

[placid lake]

ah part 2

[pine ravine]

it couldnt be

[vestal breach]

f

[pine ravine]

mine does work on edges

bottom screen shot is numbers my system caught

[woven solar]

one fun bug I can see people getting is ||accidentally parsing + before numbers as part of the number, if one carelessly uses a number-parsing utility||

[pine ravine]

why even need a number parser

[signal copper]

i don't know enough built-ins to do this pythonically, so i'll just nest for loops

[placid lake]

i bet the aoc libs have them

[pine ravine]

hi edd

[signal copper]

hi

[pine ravine]

i figure i must be either reading the instructions wrong or there is some weird edge case my brain cant think of

[signal copper]

you're gonna have to go through all of the input comparing against your output if you can't see the issue

[pine ravine]

i did 💀

[harsh obsidian]

Could somebody send me the first advent of code? (The link to it)

[earnest forge]

I mean if ||one uses a number-parsing utility|| they may also ||accidentally parse certain constructs as floats||

[lusty rover]

Tests pas
Actual does not

[lusty rover]

https://adventofcode.com/

[earnest forge]

Do you mean https://adventofcode.com/2015 or https://adventofcode.com/2023/day/1 ?

[signal copper]

then you didn't do it right 😛

[pine ravine]

agree

[woven solar]

make sure you don't count a * as a gear if it has >2 adjacent numbers, I guess.

[earnest forge]

(or <2)

[harsh obsidian]

Thank you very much!

[harsh obsidian]

😅
I meant this year XD..
Thank you very much!

[vestal kelp]

what an annoying challenge in day 3

[royal gull]

I enjoyed each of the challenges so far, because I made mistakes in each of them.

[muted aurora]

Today's one was kinda pain

All kinds of pain

[naive stone]

ahhhh i did it. that was super fun

[pine ravine]

it was them gosh dang new line characters messing me up

[regal pumice]

lol I can't get past day 1 part 2

wtf

[woven solar]

yeah, it was absurdly annoying for me too

the trick there is ||in a line "oneight", the first digit is "one" and the last digit is "eight" - they can overlap||. the example doesn't really explain this.

[strange magnet]

looking for someone to help me with my code it works just trying to see if i can make it better or should add anything

[lusty rover]

Paste it in the spoilers thread for the right day and people will take a look

[strange magnet]

okie dokie thanks

[woven sable]

wait that's a rule?

i didn't do that and i still got past it

[earnest forge]

The puzzle does specify exactly 2

[woven solar]

i don't know if it's actually enforced, but the puzzle statement is fairly unambigous

A gear is any * symbol that is adjacent to exactly two part numbers.

[woven sable]

weird

i passed both the test data and my own data

without ever doing that

i j counted a gear as anything with over 1 adjacent number

[woven solar]

smh, literally unplayable

[placid lake]

huh i counted any over 1, yeah

[woven solar]

2023 day 1: horrible edge case not explained in the statement at all
2023 day 3: tons of potential edge cases that carefully never occur in the actual inputs

[placid lake]

fr

[woven sable]

that's odd lmao

[covert arrow]

oh I completely missed that too lol

[woven sable]

i honestly j never read the prompts correctly

[earnest forge]

'tons of potential edge cases' no the puzzle's unambiguous this time it's just that you're allowed to make extra assumptions today

[woven sable]

atl it didn't set me back today

[covert arrow]

typical advent of reading comprehension

[woven sable]

💀💀

[lusty rover]

now do day 2

[woven sable]

i loved day 2

fav day so far

[woven solar]

day2 was straightforward, no complaints

[woven sable]

it was j fun to implement tbh

[placid lake]

day 2 was great

[woven solar]

woke up, did day2, went to sleep until midday, no complaints

[hollow summit]

it's either one of two solutions:

1. parsing games into a data structure
2. string manipulation

[woven sable]

im not lookinb@foreard to the puzzle that has a really easy part 1 and then ur code is too slow for part two

there's always some day like that near the beginning

ik in 2021 it was the jellyfish one

[naive stone]

i can't wait