empty tree Dec 1, 2023, 11:59 PM

#

was my part 1

#

im here but i dont want to really look at other solutions much until i figure out part 2

pseudo minnow Dec 2, 2023, 12:00 AM

#

Oooh, I have loads of issues reading very loopy stuff 😄

#

This is scala but you can more or less replace every map with a list comprehension or so:

def part1Solution(rawInput: String): Int =
  rawInput
    .split("\n")
    .map(_.filter(_.isDigit))
    .map(line => (line.head.asDigit * 10) + line.last.asDigit)
    .sum

empty tree Dec 2, 2023, 12:01 AM

#

thats super simple

pseudo minnow Dec 2, 2023, 12:02 AM

#

Second one is harder tho, especially for a day 1 💀

cosmic mica Dec 2, 2023, 12:04 AM

#

pseudo minnow Second one is harder tho, especially for a day 1 💀

Yeah agreed, especially sincve they did people dirty by replace, replacing the first occurrence and breaking the actual first word.

empty tree Dec 2, 2023, 12:05 AM

#

idk if im gonna figure out the part 2

pseudo minnow Dec 2, 2023, 12:06 AM

#

I used a hack that trivialises part 2 #1180009538544488478 message

cosmic mica Dec 2, 2023, 12:06 AM

#

@pseudo minnow ```py
print(*(v:=sorted(map(lambda s:sum(s),map(lambda x:map(int,x.split("\n")),open("inp.txt").split("\n\n")))),v[-1],sum(v[-3:]))[1:])

pseudo minnow Dec 2, 2023, 12:08 AM

#

yeah I remember that one, considerably easier 😄

cosmic mica Dec 2, 2023, 12:08 AM

#

AOC 2022 start was pretty easy until they hit us with the crane and the stacked containers 💀

#

ppl def cheated by splitting the input in 2 in the txt.

pseudo minnow Dec 2, 2023, 12:09 AM

#

It got really ugly for me there because I overengineered it

cosmic mica Dec 2, 2023, 12:10 AM

#

I wanted to get the crates with programming so this is what it turned out ```py
crates = [list(filter(lambda h: h != " ", k)) for k in list(zip(8[iter([z[1] for y in zip(*zip(9[iter(zip(4[iter(x)]))])) for z in y])]))]

#

After that it was pretty easy.

pseudo minnow Dec 2, 2023, 12:13 AM

#

Reading golfed solutions give me a headache (in a good way!) 😄

cosmic mica Dec 2, 2023, 12:20 AM

#

pseudo minnow Reading golfed solutions give me a headache (in a good way!) 😄

I got into semi-golf solutions, they do learn myself new tricks and shortcuts that I actually use in production code.

#

Not stuff like eval or bitwise shifting with binary, that's too far.

barren kindle Dec 2, 2023, 12:31 AM

#

@upper thicket two -> two2two

upper thicket Dec 2, 2023, 12:32 AM

#

barren kindle <@1005756074973990922> two -> two2two

oh my god

barren kindle Dec 2, 2023, 12:32 AM

#

lol

old prism Dec 2, 2023, 12:37 AM

#

cleaned up my solution a bit
the slice assignment keeps the lists at a max of 2 elements```py
from pathlib import Path

with open(Path(file).parent / "input") as f:
data = f.read().rstrip()

from re import finditer, Match, compile

numbers = "zero, one, two, three, four, five, six, seven, eight, nine".split(", ")

py3.12 woo

pattern = compile(fr"(?=(\d|{"|".join(numbers)}))")

p1 = p2 = 0
for line in data.splitlines():
values_1 = []
values_2 = []

for m, in map(Match.groups, finditer(pattern, line)):
    try:
        v = int(m)
    except ValueError:
        values_2[1:] = numbers.index(m),
        # values_2.append(numbers.index(m))
    else:
        values_1[1:] = values_2[1:] = v,
        # values_1.append(v)
        # values_2.append(v)

p1 += 10*values_1[0] + values_1[-1]
p2 += 10*values_2[0] + values_2[-1]

print(p1, p2)

#

though appending is probably much faster and easier to understand lol

empty tree Dec 2, 2023, 1:02 AM

#

is there a way to break up a string based on dictionary key matches

#

im trying to avoid regex. I thought just replace all instances of "one" with "1",etc

fading remnant Dec 2, 2023, 1:04 AM

#

i used subarrays

#

to test each one

#

O(N^2)

#

time complexity

regal spear Dec 2, 2023, 1:14 AM

#

Complexity is negligible since each line is pretty short

crisp sequoia Dec 2, 2023, 1:20 AM

#

my answer is incorrect yet every test case returns the right answer

#

do i have to scan al the 1000 strings ans answers now?😢

ancient cedar Dec 2, 2023, 1:22 AM

#

crisp sequoia my answer is incorrect yet every test case returns the right answer

try ||oneight||, the result should be ||18||

crisp sequoia Dec 2, 2023, 1:23 AM

#

yep it returned that

ancient cedar Dec 2, 2023, 1:24 AM

#

share your code in #aoc-solution-hints

dense roost Dec 2, 2023, 1:25 AM

#

that's what i thought at first. "it'd be trivial bc i can just ctrl + f replace everything!"
then i saw overlapping words :)

#

i gave up the mobile thing, not that masochistic lol
i'm not super well-versed in spreadsheet magic, so here's a clunky (but works!) nine nested substitutions solution:
https://docs.google.com/spreadsheets/d/1p0RXE_xMzGnM5gIPlGghr0uKjSwfVgQxdx1fWuSFDmc/edit?usp=sharing

crisp sequoia Dec 2, 2023, 1:28 AM

#

ancient cedar share your code in <#1178802982981214278>

posted it

empty tree Dec 2, 2023, 1:29 AM

#

im learning regex for this stupid thing

#

lol

barren kindle Dec 2, 2023, 1:30 AM

#

dense roost i gave up the mobile thing, not that masochistic lol i'm not super well-versed i...

I did code my thing on my phone

#

in a language I've never used before

#

not that bad at the end of the day 🥴

dense roost Dec 2, 2023, 1:36 AM

#

barren kindle I did code my thing on my phone

oh hell naw 😩 how much pain were you in

dense roost Dec 2, 2023, 1:36 AM

#

barren kindle not that bad at the end of the day 🥴

lemon_glass

barren kindle Dec 2, 2023, 1:36 AM

#

I'm used to it 😛

#

ssh to my desktop editing in vim

dense roost Dec 2, 2023, 1:37 AM

#

lemon_eyes

#

and i thought mobile spreadsheeting was bad

#

i can't imagine mobile coding

barren kindle Dec 2, 2023, 1:38 AM

#

with a decent phone keyboard it's quite fine

dense roost Dec 2, 2023, 1:38 AM

#

oh cool

#

like physical keyboard?

barren kindle Dec 2, 2023, 1:38 AM

#

no

#

on screen keyboard

#

I've sworn by https://github.com/klausw/hackerskeyboard for ages 😛

GitHub

GitHub - klausw/hackerskeyboard: Hacker's Keyboard (official)

Hacker's Keyboard (official). Contribute to klausw/hackerskeyboard development by creating an account on GitHub.

#

full qwerty keyboard

dense roost Dec 2, 2023, 1:39 AM

#

looks sweet

#

i need to try this out sometime

#

what was the random language for today?

barren kindle Dec 2, 2023, 1:40 AM

#

but overall that and then vim is actually quite alright

#

I guess in part because not a lot of mouse dependence

regal spear Dec 2, 2023, 1:40 AM

#

1seven8seven should return 17, not 18

barren kindle Dec 2, 2023, 1:40 AM

#

dense roost what was the random language for today?

groovy

dense roost Dec 2, 2023, 1:41 AM

#

barren kindle groovy

never heard of it lol

#

istg one day you'll get a ping for malbolge

barren kindle Dec 2, 2023, 1:41 AM

#

groovy is mainly used as a config language for java iirc

ancient cedar Dec 2, 2023, 1:42 AM

#

dense roost istg one day you'll get a ping for malbolge

I'm hoping for a couple of esolangs in the first few days

barren kindle Dec 2, 2023, 1:42 AM

#

(though maybe I'm thinking of something different there)

ancient cedar Dec 2, 2023, 1:42 AM

#

piet would be an interesting one

bitter copper Dec 2, 2023, 1:42 AM

#

barren kindle groovy is mainly used as a config language for java iirc

gradle?

regal spear Dec 2, 2023, 1:42 AM

#

groovy is if java was python

#

all the worst parts of python were added to java

barren kindle Dec 2, 2023, 1:44 AM

#

bitter copper gradle?

I was thinking of something that was often used with gradle, but I'm probably misremembering

#

I'm curious why groovy didn't take off, but kotlin did

#

oh wait, are gradle config files groovy?

bitter copper Dec 2, 2023, 1:47 AM

#

uhhh

#

¯_(ツ)_/¯

barren kindle Dec 2, 2023, 1:48 AM

#

Gradle scripts are written in either Groovy DSL or Kotlin DSL (domain-specific language).

#

apparently it is, and I'm not entirely making stuff up

smoky escarp Dec 2, 2023, 1:48 AM

#

dense roost that's what i thought at first. "it'd be trivial bc i can just ctrl + f replace ...

You've probably seen it by now, but if you put the word back in you can still use regex replace.

upper thicket Dec 2, 2023, 2:07 AM

#

fair glade Dec 2, 2023, 2:09 AM

#

heh

#

just finished

#

couldn't do it earlier, because of school and other work 😔

#

oh huh, I didn't even consider regex

#

that was probably the way to go 👀

#

mine is sort of bad

dense roost Dec 2, 2023, 2:15 AM

#

aboo! :D

fair glade Dec 2, 2023, 2:15 AM

#

from more_itertools import substrings_indexes

def part_1(inp: str) -> int:
    lines = inp.splitlines()
    digits = ["".join(i for i in line if i.isdigit()) for line in lines]
    return sum((int(d[0] + d[-1]) for d in digits))


def part_2(inp: str) -> int:
    lines = inp.splitlines()
    numbers = {
        "one": "1",
        "two": "2",
        "three": "3",
        "four": "4",
        "five": "5",
        "six": "6",
        "seven": "7",
        "eight": "8",
        "nine": "9",
    }

    ret = 0

    for line in lines:
        substrings = (
            "".join(i[0])
            for i in sorted(substrings_indexes(line), key=lambda t: t[1])
            if len(i) <= 5
        )
        digits = [i for i in substrings if i.isdigit() or i in numbers]

        first = digits[0] if len(digits[0]) < 2 else numbers[digits[0]]
        last = digits[-1] if len(digits[-1]) < 2 else numbers[digits[-1]]

        ret += int(first + last)

    return ret

fair glade Dec 2, 2023, 2:15 AM

#

dense roost aboo! \:D

hi 👀

dense roost Dec 2, 2023, 2:15 AM

#

i emerge from the vast dark to partake in aoc

fair glade Dec 2, 2023, 2:15 AM

#

heh

dense roost Dec 2, 2023, 2:15 AM

#

then i lurk again

fair glade Dec 2, 2023, 2:15 AM

#

😔

dense roost Dec 2, 2023, 2:16 AM

#

trying to sm cool for every early day lol

fair glade Dec 2, 2023, 2:16 AM

#

I think this year's first puzzle was harder than last year's 👀

dense roost Dec 2, 2023, 2:16 AM

#

today i spreadsheeted

#

and on mobile for part 1

fair glade Dec 2, 2023, 2:16 AM

#

I want to do my solutions in agda

dense roost Dec 2, 2023, 2:16 AM

#

what's agda?

fair glade Dec 2, 2023, 2:17 AM

#

fair glade ```py from more_itertools import substrings_indexes def part_1(inp: str) -> int...

intead of regexes I thought "windows", and turns out more_itertools has a fancy substrings function

fair glade Dec 2, 2023, 2:17 AM

#

dense roost what's agda?

https://github.com/agda/agda a programming language 👀 it's functional and dependently typed

#

you can also do proofs and stuff with it

barren kindle Dec 2, 2023, 2:29 AM

#

fair glade ```py from more_itertools import substrings_indexes def part_1(inp: str) -> int...

that's not groovy

bitter copper Dec 2, 2023, 2:30 AM

#

yeah what's up with that anyway

drowsy kelp Dec 2, 2023, 2:34 AM

#

Was stumped about the overlapping numbers like a lot of people, before watching hyper-nutrino's video and realising the problem, as I am a noob, any obvious optimizations or things that indicate I don't know about some feature that I should?


import regex as re

with open('day1/input.txt','r') as f:
    inputlines = f.read().splitlines()

def part1(inputlines):
    ans = 0
    for line in inputlines:
        numline = []
        for char in line:
            if char.isdigit():
                numline.append(int(char))
        print(numline[-1])
        ans += (int(f'{numline[0]}{numline[-1]}'))

    return ans

def part2(inputlines):
  
    wordstonum = {
        'one':'1',
        'two':'2',
        'three':'3',
        'four':'4',
        'five':'5',
        'six':'6',
        'seven':'7',
        'eight':'8',
        'nine':'9'
    }

    def num(x: str):
        if x.isdigit():
            return x
        return wordstonum[x]
    
    pattern = "(?=(" + "|".join(wordstonum.keys()) + "|\\d))"
    ans = 0

    for line in inputlines:
        digits = re.findall(pattern,line)
        digits = [*map(num,digits)]
        ans += (int(f'{digits[0]}{digits[-1]}'))

    return (ans)

print(part1(inputlines))
print(part2(inputlines))```

mortal siren Dec 2, 2023, 2:35 AM

#

drowsy kelp Was stumped about the overlapping numbers like a lot of people, before watching ...

Is this code working for you?

fair glade Dec 2, 2023, 2:37 AM

#

barren kindle that's not groovy

I didn't join the roulette leaderboard 🥺

#

I'll still do a solution in groovy, just after the other handful of languages on my list to do the solutions in

drowsy kelp Dec 2, 2023, 2:41 AM

#

mortal siren Is this code working for you?

Yup!

urban saddle Dec 2, 2023, 2:43 AM

#

I'm quite happy with this solution. A functional and recursive approach to doing Day 1 in Elixir. https://github.com/soupglasses/advent-of-code/blob/main/2023/day_01.exs

GitHub

advent-of-code/2023/day_01.exs at main · soupglasses/advent-of-code

My personal advent of code repo. Contribute to soupglasses/advent-of-code development by creating an account on GitHub.

flint hazel Dec 2, 2023, 2:45 AM

#

with open("day1.txt") as f:
  puzzle = f.read().split("\n")

wordsToDigit = {
  "one": "o1e",
  "two": "t2o",
  "three": "t3e",
  "four": "f4r",
  "five": "f5e",
  "six": "s6x",
  "seven": "s7n",
  "eight": "e8t",
  "nine": "n9e"
}

digits = []
windowSize = 5
words = wordsToDigit.keys()
for i in puzzle:
  iterations = [i]
  for n in range(len(i) - windowSize + 1):
    smallS = iterations[-1][n: n+windowSize]
    for word in words:
      if word in smallS:
        i = i.replace(word, str(wordsToDigit[word]), 1)
        iterations.append(i)
  digitized = "".join(filter(str.isdigit, iterations[-1]))
  digits.append(int(f"{digitized[0]}{digitized[-1]}"))

print(sum(digits))```
tunnel vision: the puzzle

mortal siren Dec 2, 2023, 2:53 AM

#

drowsy kelp Yup!

It is curious that you use third-party regex without the main reason to use it in this day's puzzle :)

#

But regardless the solution looks clean

fair glade Dec 2, 2023, 2:56 AM

#

fair glade I want to do my solutions in agda

oohkay, that might not happen anytime soon

#

from what I can tell, editor support for agda isn't amazing

#

at least, in terms of what I'm used to

drowsy kelp Dec 2, 2023, 3:06 AM

#

mortal siren It is curious that you use third-party regex without the main reason to use it i...

overlapped=True? Yeah this was my first time using regex and I have no idea what I'm doing so I assumed regex and re were the same thing, didn't even realise I had the non built-in module until I read other people's solutions :P

pure widget Dec 2, 2023, 3:15 AM

#

crimes (groovy solution)

def part1(lines) {
    lines.sum {
        def digits = it.findAll(/\d/).collect { it.toInteger() }
        digits[0] * 10 + digits[-1]
    }
}

def part2(lines) {
    def wordForm = ["", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine"]
    def fore_regex = /\d|one|two|three|four|five|six|seven|eight|nine/
    def back_regex = /\d|eno|owt|eerht|ruof|evif|xis|neves|thgie|enin/
    lines.sum {
        def first = it.find(fore_regex)
        def first_num = first.isNumber() ? first.toInteger() : wordForm.indexOf(first)
        def second = it.reverse().find(back_regex)
        def second_num = second.isNumber() ? second.toInteger() : wordForm.indexOf(second.reverse())
        first_num * 10 + second_num
    }
}

File input = new File("inputs/day1.txt")
def lines = input.readLines()
println "part 1: ${part1 lines}"
println "part 2: ${part2 lines}"

hasty void Dec 2, 2023, 3:16 AM

#

pure widget crimes (groovy solution) ```groovy def part1(lines) { lines.sum { de...

what language is this?

pure widget Dec 2, 2023, 3:16 AM

#

Groovy, the language roulette of the day (see the heading under #announcements message)

hasty void Dec 2, 2023, 3:16 AM

#

the syntax is interesting

hasty void Dec 2, 2023, 3:17 AM

#

pure widget Groovy, the language roulette of the day (see the heading under https://canary.d...

ooh i see I did not know that was a thing

#

I'm too committed to doing it in my own, but sounds fun!

#

can i suggest my language as part of the roulette? 👀

pure widget Dec 2, 2023, 3:18 AM

#

you'd have to ask kat

#

I think they're all slightly more well-known ones though, and I believe the schedule is set?

bitter copper Dec 2, 2023, 3:18 AM

#

source?

fair glade Dec 2, 2023, 3:19 AM

#

your mother

#

dawn I'm having a crisis

#

agda-mode is making me do emacs shortcuts in vscode

pure widget Dec 2, 2023, 3:24 AM

#

lovely

pure widget Dec 2, 2023, 3:24 AM

#

bitter copper source?

no.

bitter copper Dec 2, 2023, 3:43 AM

#

pure widget no.

i disagree

barren kindle Dec 2, 2023, 4:06 AM

#

pure widget Groovy, the language roulette of the day (see the heading under https://canary.d...

did you join the roulette leaderboard? 👀

pure widget Dec 2, 2023, 4:06 AM

#

uhhh no, i was not aware there was one

uneven lava Dec 2, 2023, 4:06 AM

#

it's up there somewhere, I have to pin it still

bitter copper Dec 2, 2023, 4:07 AM

#

I'm getting clowned on in the other leaderboards I'm in

barren kindle Dec 2, 2023, 4:07 AM

#

we be growing

bitter copper Dec 2, 2023, 4:07 AM

#

but I don't wanna stay up late

barren kindle Dec 2, 2023, 4:08 AM

#

to me it's just nice to see collective progress

barren kindle Dec 2, 2023, 4:11 AM

#

pure widget uhhh no, i was not aware there was one

#advent-of-code message

pure widget Dec 2, 2023, 4:11 AM

#

ty

bitter copper Dec 2, 2023, 4:13 AM

#

half the time I spent today was installing the language ngl

pure widget Dec 2, 2023, 4:13 AM

#

nix ftw

south badger Dec 2, 2023, 4:14 AM

#

bitter copper half the time I spent today was installing the language ngl

I already had sdkman installed from.... scala IIRC, so I just pasted in "sdkman install groovy", and away it went

#

Done in like three minutes

#

And then it took me 90 minutes to get a working solution firRip

pure widget Dec 2, 2023, 4:15 AM

#

lemon_fingerguns_shades https://github.com/dawnofmidnight/aoc-2023/blob/main/flake.nix#L14

crude ledgeBOT Dec 2, 2023, 4:15 AM

#

flake.nix line 14

groovy # day 1```

bitter copper Dec 2, 2023, 4:17 AM

#

neat

#

it may be devcontainer time

fair glade Dec 2, 2023, 4:30 AM

#

Agda took like 20 minutes to install ;-;

keen gulch Dec 2, 2023, 4:30 AM

#

pure widget <:lemon_fingerguns_shades:754441879960944641> <https://github.com/dawnofmidnight...

I started using nixos as my main OS. Having never done functional programming like all you CS majors (presumably), I'm still lost by the nix language itself. But having every install and every system change being 100% undoable is worth it even if I'm not getting the most out of it yet.

fair glade Dec 2, 2023, 4:31 AM

#

my intention is to nuke my fedora install, reinstall, and then set up everything using nix 👀

pure widget Dec 2, 2023, 4:31 AM

#

Do you want some resources that helped me with the language?

fair glade Dec 2, 2023, 4:31 AM

#

I looked into silverblue, but I’m not sure how well that plays with dual booting

keen gulch Dec 2, 2023, 4:34 AM

#

pure widget Do you want some resources that helped me with the language?

I have https://nixcloud.io/tour/?id=1 on my list of things to go through as the nix discord channel recommended it since I wanted something with examples I could practice with, but if you have other resources that you liked. I went through the nix.dev manual and, I donno, some parts just aren't clicking and I think its because I'm just not used to functional languages.

pure widget Dec 2, 2023, 4:34 AM

#

Understandable. https://nixos.org/guides/nix-pills/ is also a nice digestible series of really short blog-post style articles, a couple of which touch on the language.

#

The real pain is working with nixpkgs and writing derivations 😅

#

I'm decent at the language at this point, but that is just beyond me for anything nontrivial

fair glade Dec 2, 2023, 4:36 AM

#

are you fully nixed, dawn?

pure widget Dec 2, 2023, 4:36 AM

#

Nah, I'm using it in wsl

fair glade Dec 2, 2023, 4:36 AM

#

I see

pure widget Dec 2, 2023, 4:37 AM

#

still need some software that exists only on or is better on windows

keen gulch Dec 2, 2023, 4:38 AM

#

I started going through the pills at one point, but when I asked some follow up questions on the nix discord about nix-env, they told me thats out dated and not to use pills because its focus on an older style 🤷‍♂️ . My whole system configuration is in a flake using home manager, so it seems like there are a lot of resources that aren't really targetted to a newer setup like that

fair glade Dec 2, 2023, 4:38 AM

#

I dual boot, which is nice with a drive dedicated each to windows and Linux

keen gulch Dec 2, 2023, 4:39 AM

#

keen gulch I started going through the pills at one point, but when I asked some follow up ...

Why did Sir Lancebot react?

pure widget Dec 2, 2023, 4:39 AM

#

The word "flake"

fair glade Dec 2, 2023, 4:39 AM

#

Christmas reactions

pure widget Dec 2, 2023, 4:39 AM

#

Huh, interesting. I also use a Nix flake config, although it's quite minimal because I use WSL, as I said

#

I used to not know the home-manager option search existed, that made my config so much cleaner

#

oh woops, we've gotten wildly off topic

dense roost Dec 2, 2023, 4:51 AM

#

christmas

#

flake

#

pretty cool

grave shale Dec 2, 2023, 7:57 AM

#

gpt4 can't get part 2 right!

upbeat eagle Dec 2, 2023, 8:21 AM

#

i decided to give it a shot with dumb approaches. for whatever reason, i decided to check the str from both directions to get short circuiting, but then did the checks in the dumbest ways possible 😌

#

accumulator = 0

templates = {"o": ["one"],
             "t": ["two", "three"],
             "f": ["four", "five"],
             "s": ["six", "seven"],
             "e": ["eight"],
             "n": ["nine"]}
mappings = {"one": "1", "two": "2", "three": "3", "four": "4", "five": "5", "six": "6", 
            "seven": "7", "eight": "8", "nine": "9",}

def check_digit(index, line, templates, mappings):
    total_length = len(line)
    char = line[index]

    if char.isdigit():
        return char

    inventory = templates.get(char, [])
    for item in inventory:
        
        item_length = len(item)
        if index + item_length - 1 < total_length:
            chunk = line[index : index + item_length]
            number = mappings.get(chunk)
            if number is not None:
                return number
    
    return "F" # to pay respects

with open("inputs/input_day1_.txt", "r") as file:
    for line in file:
        number = ""

        n = 0
        while n < len(line):
            digit = check_digit(n, line, templates, mappings)
            if digit.isdigit():
                number = digit
                break
            n += 1
        
        n = len(line) - 1
        while n >= 0:
            digit = check_digit(n, line, templates, mappings)
            if digit.isdigit():
                number += digit
                break
            n -= 1

        accumulator += int(number)
        
print(accumulator)

scenic jay Dec 2, 2023, 8:25 AM

#

    return "F" # to pay respects

i'm stealing this

crimson coral Dec 2, 2023, 9:07 AM

#

Where can I get help with my day 1 solution?

oak epoch Dec 2, 2023, 9:11 AM

#

crimson coral Where can I get help with my day 1 solution?

Make a thread in #aoc-solution-hints and I might pop over and help

crisp sequoia Dec 2, 2023, 2:54 PM

#

finally did it ```py
import aoc_lube
import re
LINES = aoc_lube.fetch(year=2023, day=1).splitlines()

LINES = ['1abc2', 'pqr3stu8vwx', 'a1b2c3d4e5f', 'treb7uchet']

lines = 0
index = 0
total_number = 0
ValueError_count = 0
nummers = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
spelled_nummers = {
'zero': 0,
'one': 1,
'two': 2,
'three': 3,
'four': 4,
'five': 5,
'six': 6,
'seven': 7,
'eight': 8,
'nine': 9,
}
reverse_spelled_nummers = {
0: 'zero',
1: 'one',
2: 'two',
3: 'three',
4: 'four',
5: 'five',
6: 'six',
7: 'seven',
8: 'eight',
9: 'nine',
}

print(LINES)

for string in LINES:
index_list = []
index_library = {}
print(string)
for number in spelled_nummers.keys():
if number in string:
indexes = [m.start() for m in re.finditer(number, string)]
for item in indexes:
index_library[item] = spelled_nummers.get(number)

for number in reverse_spelled_nummers.keys():
    if str(number) in string:
        print(str(number))
        indexes = [m.start() for m in re.finditer(str(number), string)]
        for item in indexes:
            index_library[item] = number

print(index_library)
index_list = []
for index in index_library.keys():
    index_list.append(index)
index_list.sort()
print(index_list)
first_number = index_library.get(index_list[0])
last_number = index_library.get(index_list[-1])
number_to_add = int(f'{first_number}{last_number}')
total_number = total_number + number_to_add
print(number_to_add)

print(total_number)

woeful gazelle Dec 2, 2023, 2:55 PM

#

mappings = {
    "one": "1",
    "two": "2",
    "three": "3",
    "four": "4",
    "five": "5",
    "six": "6",
    "seven": "7",
    "eight" : "8",
    "nine": "9",
}
with open('day1.txt') as puzzle_file:
    for line in puzzle_file:
        line = line.strip()

        reduce_one = reduce(lambda a, kv: a.replace(*kv), mappings.items(), line)
        reduce_two = reduce(lambda a, kv: a.replace(*kv), reversed(list(mappings.items())), line)

        numbers = []
        final, final_reversed = zip(reduce_one, reduce_two), zip(reversed(reduce_one), reversed(reduce_two))

        for r1, r2 in final:
            if r1.isdigit():
                sum += int(r1)*10
                break
            if r2.isdigit():
                sum += int(r2)*10
                break

        for r1, r2 in final_reversed:
            if r2.isdigit():
                sum += int(r2)
                break
            if r1.isdigit():
                sum += int(r1)
                break

Non-Regex solution (because I forgot regex existsted)

#

Basically, just using .replace() twice - first time it replaces going down from 10->9->8 etc, second time 1->2->3

graceful trail Dec 2, 2023, 3:13 PM

#

import re
f = open("input.txt", "r")

digit_pattern = r"\d|one|two|three|four|five|six|seven|eight|nine"

cor = {"one": "1", "two": "2", "three": "3", "four": "4", "five": "5", "six": "6", "seven": "7", "eight": "8", "nine": "9"}

items = f.readlines()
_sum = 0
for item in items:
    reg = re.search(fr".*?({digit_pattern}).*({digit_pattern}).*", item)
    nb = None
    if reg is None:
        reg = re.search(fr".*?({digit_pattern}).*", item).group(1)
        nb = cor.get(reg, reg)*2
    else:
        nb = "".join(cor.get(i, i) for i in reg.groups())
    _sum += int(nb)
print(_sum)

f.close()

``` everyone seems to be using findall, didn't even came to my mind

lime hatch Dec 2, 2023, 4:21 PM

#

I aspire to write as clean code. Ultra performative

silk oracle Dec 2, 2023, 4:47 PM

#

Raku:

my @lines = "day_01.input".IO.lines;
## Part 1
@lines>>.comb(/\d/)>>.[0, *-1]>>.join.sum.put;

## Part 2
my %digs = <one two three four five six seven eight nine> Z=> 1..9;
# Fast to write, slow to execute
@lines>>.match(/ \d | @(%digs.keys) /, :ex)>>.[0, *-1]>>.map({%digs{$_} // $_})>>.join.sum.put;
# Slow to write, fast to execute
@lines.map(-> \ℓ { (.min({.value.min}).key, .max({.value.max}).key).join given (%digs{$_} // $_ => ℓ.indices($_) for %digs.kv) }).sum.put;

pure widget Dec 2, 2023, 5:11 PM

#

initial d1p1 solution

l←⊃⎕NGET'./day1.txt' 1
⎕←+/{⍎v[1,≢v←⍵/⍨⍵∊⎕d]}¨l

bitter bluff Dec 2, 2023, 8:02 PM

#

nums = {
    "one": 1,
    "two": 2,
    "three": 3,
    "four": 4,
    "five": 5,
    "six": 6,
    "seven": 7,
    "eight": 8,
    "nine": 9,
}

with open("aoc/2023/input.txt") as file:
    total = 0
    for line in file.readlines():
        ans = []
        for iteration in (enumerate(line), reversed(list(enumerate(line)))):
            for i, c in iteration:
                if c.isdigit():
                    ans.append(int(c))
                    break
                for numk, num in nums.items():
                    if line[i:].startswith(numk):
                        ans.append(num)
                        break
                else:
                    continue
                break
        total += ans[0] * 10 + ans[1]

    print(total)

shitass solution i spent an hour on because i was not breaking out of a nested loop, just remembered for else exists pithink

mortal siren Dec 2, 2023, 8:03 PM

#

for-else exists but using an early return in a function is much nicer (and can break out of nested loops easily)

cosmic mica Dec 2, 2023, 8:52 PM

#

Who loves parsing?

limber pulsar Dec 3, 2023, 12:15 AM

#

I am actually really salty that this was the issue

#

Also somewhat disappointed in the fact that python's builtin re doesnt' have overlapping matches at all

mortal siren Dec 3, 2023, 12:39 AM

#

limber pulsar I am actually really salty that this was the issue

I’m quite annoyed this is a thing too

limber pulsar Dec 3, 2023, 12:40 AM

#

Normally AOC is good about catching these technical issues that you're not really expected to be searchign for on day one in the example...

mortal siren Dec 3, 2023, 12:40 AM

#

limber pulsar Also somewhat disappointed in the fact that python's builtin re doesnt' have ove...

They were meant to replace re with regex years ago - seems like that never happened

mortal siren Dec 3, 2023, 12:41 AM

#

limber pulsar Normally AOC is good about catching these technical issues that you're not reall...

I was aware of the issue but I thought it wouldn’t count. I thought if a letter makes up a number word which counts as a “digit” then you can’t reuse it

mortal siren Dec 3, 2023, 12:41 AM

#

mortal siren They were meant to replace re with regex years ago - seems like that never happe...

Probably for good reasons, I should add

limber pulsar Dec 3, 2023, 12:41 AM

#

I thought python's regex module did overlapping

#

I'm also sleep deprived

mortal siren Dec 3, 2023, 12:41 AM

#

limber pulsar I thought python's regex module did overlapping

Not re. Only third party regex

limber pulsar Dec 3, 2023, 12:42 AM

#

AOC should have caught it
Python should've fixed it by now
I should have noticed

#

sigh

mortal siren Dec 3, 2023, 12:42 AM

#

Well I guess it depends if you count ?= as overlapping support or not

limber pulsar Dec 3, 2023, 12:42 AM

#

I'd call that a hack

mortal siren Dec 3, 2023, 12:42 AM

#

In some sense it does support it but not in an obvious way

#

Yeah that’s fair

limber pulsar Dec 3, 2023, 12:43 AM

#

Not like whatever but

#

If I had to do that in production code

#

I'd comment it with #hack required because _insert reason here_

mortal siren Dec 3, 2023, 12:43 AM

#

Yeah indeed

#

I try not to use crazy regex in prod code

regal merlin Dec 3, 2023, 12:43 AM

#

# hack required for job security trollface

mortal siren Dec 3, 2023, 12:44 AM

#

I don’t want to be that guy whose Python looks like Morse code lol

limber pulsar Dec 3, 2023, 12:44 AM

#

I hacked some django theming code with

#

# I know why this was written this way, but it's wrong. 
# So we actually do it differently. This is noted in the project readme, since the library documentation is now wrong.
# Speeds up page load times by 50%

mortal siren Dec 3, 2023, 12:44 AM

#

lol 😂

limber pulsar Dec 3, 2023, 12:45 AM

#

The actual technical reasons is a conflict between two different use cases for the theming

#

between different themes for different pages and different theme "layers"

mortal siren Dec 3, 2023, 12:46 AM

#

I do love coming across comments like that

limber pulsar Dec 3, 2023, 12:46 AM

#

the engine didn't actually support both at once so I hacked it to

#

moved on

mortal siren Dec 3, 2023, 12:46 AM

#

A classic I enjoy is a comment saying “this is temporary”

#

You just know that code is going to be there for the next 50 years

limber pulsar Dec 3, 2023, 12:48 AM

#

One of my favorite comments is

#

"this is technically like O(n^5) but it never runs on more than 15 items or outside of debug mode"

mortal siren Dec 3, 2023, 12:49 AM

#

😂

limber pulsar Dec 3, 2023, 12:51 AM

#

Actually solving the problem would be mentally hard and well, debug mode, not too slow, time to move on and whistle

#

it's part of a table layout engine

#

kind of proud of it, it makes pretty tables

#

If I were to go back and owrk on this code

#

I would probably un-nest it a bit and make the comments longer

#

code's 4 years old now

mortal siren Dec 3, 2023, 12:54 AM

#

We’re probably getting off topic for the channel, btw. But yeah, who cares about performance with small inputs - certainly not me :)

limber pulsar Dec 3, 2023, 12:55 AM

#

lol

fluid ivy Dec 3, 2023, 1:02 AM

#

Are non-Python solutions alright for this channel?

#

For part 1 I did
|| ```
Ax: input
Bx: =REGEX(Ax, "[0-9]")
Cx: =REGEX(Ax, "0-9")
Dx: =Bx*10 + Cx
E1: =SUM(D1:D99999)

For part 2 I did
|| ```
Ax: input
Bx: =SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(SUBSTITUTE(Ax, "nine", "n9e"), "eight", "e8t"), "seven", "s7n"), "six", "s6x"), "five", "f5e"), "four", "f4r"), "three", "t3e"), "two", "t2o"), "one", "o1e")
Cx: =REGEX(Bx, "[0-9]")
Dx: =REGEX(Bx, "[0-9](?!.*[0-9])")
Ex: =Cx*10 + Dx
G2: =SUM(E1:E99999)
``` ||

limber pulsar Dec 3, 2023, 1:06 AM

#

excel?

fluid ivy Dec 3, 2023, 1:06 AM

#

yeah

limber pulsar Dec 3, 2023, 1:06 AM

#

brave

fluid ivy Dec 3, 2023, 1:07 AM

#

idk why all the solutions are so long ducky_concerned maybe I am missing something

#

Basically the reason a naive .replace() chain doesn't work is overlapping digits. So something likeeightwo wouldn't work (it can't be an input because it wouldn't fit day 1, but whatever).
But if you replace eight with e8t instead of just 8, you get e8two which will work just fine

fluid ivy Dec 3, 2023, 1:10 AM

#

upper thicket

ok this one is even better and covers more edge cases theoretically (like if two letters can overlap, imagine if woops was a number and it overlapped with two)

south badger Dec 3, 2023, 2:58 AM

#

fluid ivy For part 1 I did || ``` Ax: input Bx: =REGEX(Ax, "[0-9]") Cx: =REGEX(Ax, "[0-9](...

Why ||substitute||?
Just ||check the indexes||
https://github.com/shenanigansd/scratchpad/tree/main/events/advent_of_code/2023/01/excel

#

Wait
Wait wait wait wait wait wait

Excel has REGEX !?!?

scenic jay Dec 3, 2023, 3:06 AM

#

regexcel

fading axle Dec 3, 2023, 4:36 AM

#

limber pulsar I thought python's regex module did overlapping

you can fake it with lookahead assertions

#

does kinda suck it's not builtin

plucky sentinel Dec 3, 2023, 7:40 AM

#

gaaaahhhhhhhhhhhhhh

fluid ivy Dec 3, 2023, 9:21 AM

#

south badger Why ||substitute||? Just ||check the indexes|| <https://github.com/shenanigansd/...

I'm using LibreOffice Calc so it doesn't work for me

vocal pecan Dec 3, 2023, 10:02 AM

#

What is DIGIT_MAP?

carmine ibex Dec 3, 2023, 11:37 AM

#

i don't know what im doing anymore ```py
import re
import functools

with open("./input.txt") as file:
input = file.read().strip()

nums = lambda input: sum(
int(n[0] + n[-1]) for n in re.sub(r"[a-z]", "", input).split("\n")
)
nums = lambda input: sum(
int((match := re.findall(r"(\d)", line))[0] + match[-1])
for line in input.split("\n")
)
nums = lambda input: sum(
int((s := [*filter(str.isdigit, l)])[0] + s[-1]) for l in input.split("\n")
)

print(nums(input))
print(
nums(
functools.reduce(
lambda v, ie: v.replace((e := ie[1]), e + str(ie[0] + 1) + e),
enumerate("one two three four five six seven eight nine".split()),
input,
)
)
)

mortal siren Dec 3, 2023, 12:02 PM

#

carmine ibex i don't know what im doing anymore ```py import re import functools with open("...

does this work?

carmine ibex Dec 3, 2023, 12:03 PM

#

they do

fluid ivy Dec 3, 2023, 12:56 PM

#

carmine ibex i don't know what im doing anymore ```py import re import functools with open("...

well actually you almost never want str.isdigit, use str.isdecimal instead

#

str.isdigit matches a whole bunch of characters that are not 0123456789

carmine ibex Dec 3, 2023, 12:57 PM

#

Oh?

#

I did not know that

fluid ivy Dec 3, 2023, 12:59 PM

#

carmine ibex Oh?

!e

print("௫".isdigit())

crude ledgeBOT Dec 3, 2023, 12:59 PM

#

@fluid ivy :white_check_mark: Your 3.12 eval job has completed with return code 0.

True

fluid ivy Dec 3, 2023, 12:59 PM

#

!charinfo ௫

crude ledgeBOT Dec 3, 2023, 12:59 PM

#

Character Info

\u0beb : TAMIL DIGIT FIVE - ௫

fluid ivy Dec 3, 2023, 12:59 PM

#

same with \d in regex

carmine ibex Dec 3, 2023, 1:00 PM

#

I see.

hidden schooner Dec 3, 2023, 1:12 PM

#

me using .isdigit

#

aniblobsweat

fluid ivy Dec 3, 2023, 1:19 PM

#

I imagine thousands of projects have some inconsistency because of using \d in regex

hidden schooner Dec 3, 2023, 1:21 PM

#

!e ```py
import re

print(re.search(r"\d+", "x௫d"))

crude ledgeBOT Dec 3, 2023, 1:21 PM

#

@hidden schooner :white_check_mark: Your 3.12 eval job has completed with return code 0.

<re.Match object; span=(1, 2), match='௫'>

hidden schooner Dec 3, 2023, 1:21 PM

#

dam

peak mirage Dec 4, 2023, 1:21 PM

#

fluid ivy well actually you almost never want `str.isdigit`, use `str.isdecimal` instead

oh thats bad

vocal robin Dec 4, 2023, 4:24 PM

#

import re


with open('input.txt') as f:
    calibration_document = f.read().splitlines()


def part1():
    calibration_values = []
    for line in calibration_document:
        digits = re.findall(r'\d', line)
        calibration_value = int(digits[0] + digits[-1])
        calibration_values.append(calibration_value)

    print(sum(calibration_values))


def part2():
    digit_map = {
        'one': 'o1e',
        'two': 't2o',
        'three': 't3e',
        'four': 'f4r',
        'five': 'f5e',
        'six': 's6x',
        'seven': 's7n',
        'eight': 'e8',
        'nine': 'n9e'
    }
    calibration_values = []
    for line in calibration_document:
        for digit in digit_map:
            line = line.replace(digit, digit_map[digit])

        digits = re.findall(r'\d', line)
        calibration_value = int(digits[0] + digits[-1])
        calibration_values.append(calibration_value)

    print(sum(calibration_values))


if __name__ == '__main__':
    part1()
    part2()

#

ignore overlap issue by just letting it hang out, ya feel?>

#

it looks so dumb but it works

tired zenith Dec 4, 2023, 4:26 PM

#

That's so powerfully cursed

#

I hate it and love it

vocal robin Dec 4, 2023, 4:26 PM

#

beats doing the ugly ass initial solution i had

tired zenith Dec 4, 2023, 4:28 PM

#

Fair

vocal pecan Dec 4, 2023, 5:07 PM

#

import re


file = open('input.txt', 'r')
input = [i for i in file.read().strip('\n').split('\n')]

def part1(lines):
    total = 0
    for line in lines:
        numbers = re.findall(r'[0-9]', line)
        if len(numbers) == 1:
            double = f'{numbers[0]}{numbers[0]}' 
            total += int(double)
        else:
            total += int(numbers[0] + numbers[-1])
    return total

def part2(lines):
    wordstonum = {
        'one':'1',
        'two':'2',
        'three':'3',
        'four':'4',
        'five':'5',
        'six':'6',
        'seven':'7',
        'eight':'8',
        'nine':'9'
    }

    def num(x: str):
        if x.isdigit():
            return x
        return wordstonum[x]
    
    pattern = "(?=(" + "|".join(wordstonum.keys()) + "|\\d))"
    ans = 0

    for line in lines:
        digits = re.findall(pattern,line)
        digits = [*map(num,digits)]
        ans += (int(f'{digits[0]}{digits[-1]}'))

    return (ans)


print(part2(input))
print(part1(input))

barren kindle Dec 4, 2023, 6:28 PM

#

limber pulsar Also somewhat disappointed in the fact that python's builtin re doesnt' have ove...

overlapping matches has a tendency to blow up

#

e.g.

re.findall('a+', 'a'*10000)

#

and I suspect zero length matches complicate things as well

#

(it's kinda breaking how regex works)

dreamy edge Dec 5, 2023, 12:36 AM

#

Took me 3 hours to solve day 1 and man, I made a lot of mistakes. You can watch my dumbass on stream trying to tackle it here: https://www.youtube.com/watch?v=QbmDmoC1PS0

Or be spoiled and see the solution. Do note that the code is now broken for Part 1 of Day 1. I might reorganize it to accommodate Part 1 and Part 2 solution in a single Python file. And looking at other people's code made me realize how overcomplicated my code is. Dammit. https://github.com/Qoyyuum/adventofcode/blob/2023/2023/01/trebuchet.py

YouTube

Programmer Q Dev

Advent of Code Catch Up

Powered by Restream https://restream.io

Coding late nights demands coffee. Buy me some at https://www.buymeacoffee.com/qoyyuum

You can join along or tackle the code challenges at https://adventofcode.com

The Github code repo for Advent of Code:
https://github.com/Qoyyuum/adventofcode

#Python #AdventOfCode #AOC

▶ Play video

GitHub

adventofcode/2023/01/trebuchet.py at 2023 · Qoyyuum/adventofcode

Advent of Code to all of the daily puzzles of each year. - Qoyyuum/adventofcode

fluid ivy Dec 5, 2023, 10:24 AM

#

vocal robin ```py import re with open('input.txt') as f: calibration_document = f.read...

I did exactly the same lol #1180009538544488478 message

open bay Dec 6, 2023, 9:24 AM

#

My Day1/Part 1
And youdont want to know how long it took me. a very Looooooooooooooooooooooooooooooooooooooonnnnnnnnnnnnnnnnnnnngggggggggggggggggggggggggggg time indeed. Though there were lot of iterations till arriving here. Thank you stackoverflow.

||```py
def get_integers(string):
# parse our string for integers
return [int(c) for c in string if c.isdigit()]

def count_sums():
sums = []
# loop through our data
result = [get_integers(item) for item in data]
for z in result:
# if we have more than one integer, we need to combine them
if len(z) > 1:
x = int(f"{z[0]}{z[-1]}")
# add our sum to our list
sums.append(x)
else:
# if we only have one integer, we need to double it
x = int(f"{z[0]}{z[0]}")
sums.append(x)
return sums

def main():
print(sum(count_sums()))

if name == "main":
main()

mortal siren Dec 6, 2023, 10:05 AM

#

open bay My Day1/Part 1 And youdont want to know how long it took me. a very Looooooooooo...

Congrats! Even if it took you long, nice job for sticking with it

#

Regarding your code, a style point - try to avoid naming variables as _ if you plan to use them. _ is a convention to label only variables that would be thrown away.

#

I have other code suggestions if you want to hear them :)

open bay Dec 6, 2023, 10:09 AM

#

mortal siren Congrats! Even if it took you long, nice job for sticking with it

Allways open to a critiqu. Thats how you improve. I realised my mistake with the _ now, its a throwaway and i kept using it. should have used a name. what other points

mortal siren Dec 6, 2023, 10:12 AM

#

open bay Allways open to a critiqu. Thats how you improve. I realised my mistake with the...

Another point is that you don’t actually need the if statement, because even if the length of the list is 1, you can still access the same element through index 0 or index -1

#

Last comment is that count_sums and the sums list could be renamed because you aren’t actually summing the values until later.

open bay Dec 6, 2023, 10:15 AM

#

so count_sums and sums should be different names for readability of code?
i dont see how i could have evaluated teh single digit or double digit unless i had the iff. can you show me what you mean?

scenic jay Dec 6, 2023, 10:17 AM

#

open bay so count_sums and sums should be different names for readability of code? i dont...

which element is _[-1] if _ is a single element list?

open bay Dec 6, 2023, 10:20 AM

#

I am appending a zero to the number if the length of the _ is 1, ( x = int(f"{[0]}{[-1]}"))
But I see your point having re-examined the actual input text, there are no single digits like there were in the test data.

scenic jay Dec 6, 2023, 10:25 AM

#

wdym appending a zero?

open bay Dec 6, 2023, 10:27 AM

#

im getting confused with day 4, scratchcards. Methinks comments might have been helpful

mortal siren Dec 6, 2023, 10:32 AM

#

This is AOC - no one writes comments :P

open bay Dec 6, 2023, 10:35 AM

#

haha, I just did. Ok so now that I know what im talking about and not scratchcards, I use the if to either output the number if its more than a single digit geting the fist and last figit from any number {0} or {-1} and if it is a single digit I convert it to two digits [0][0]

mortal siren Dec 6, 2023, 12:18 PM

#

open bay haha, I just did. Ok so now that I know what im talking about and not scratchcar...

But you can do it the same way

#

You can do 0 and -1 for both cases

scenic jay Dec 6, 2023, 12:27 PM

#

mortal siren This is AOC - no one writes comments :P

hey I wrote comments for day 5

mortal siren Dec 6, 2023, 12:43 PM

#

:O
I need to look at the code for that

#

I’ll do that now even though I don’t have my IDE with me :(

gray timber Dec 8, 2023, 12:33 PM

#

file = open("day1_input.txt", "r")
cum_sum = 0
number_words = {
    "one": "1",
    "two": "2",
    "three": "3",
    "four": "4",
    "five": "5",
    "six": "6",
    "seven": "7",
    "eight": "8",
    "nine": "9"
}
window_size = 5


def check_for_replace(line_to_replace):
    for word in number_words.keys():
        line_to_replace = line_to_replace.replace(word, number_words[word])
    return line_to_replace


def check_word(word_to_check):
    for i in range(len(word_to_check) - window_size + 1):
        x = check_for_replace(word_to_check[i:i + window_size])
        if x != word_to_check[i:i + window_size]:
            word_to_check = word_to_check.replace(word_to_check[i:i+window_size], x)
    return word_to_check


for line in file:
    line = check_word(line)
    nums = [i for i in line if i.isdigit()]
    line_value = int(nums[0] + nums[-1])
    cum_sum += line_value
    break
print(cum_sum)

Does anyone know where i went wrong?

mortal siren Dec 8, 2023, 2:06 PM

#

gray timber ```py file = open("day1_input.txt", "r") cum_sum = 0 number_words = { "one":...

have you considered the oneight case?

gray timber Dec 8, 2023, 2:15 PM

#

yeah

#

!e

#file = open("day1_input.txt", "r")
cum_sum = 0
number_words = {
    "one": "1",
    "two": "2",
    "three": "3",
    "four": "4",
    "five": "5",
    "six": "6",
    "seven": "7",
    "eight": "8",
    "nine": "9"
}
window_size = 5


def check_for_replace(line_to_replace):
    for word in number_words.keys():
        line_to_replace = line_to_replace.replace(word, number_words[word])
    return line_to_replace


def check_word(word_to_check):
    for i in range(len(word_to_check) - window_size + 1):
        x = check_for_replace(word_to_check[i:i + window_size])
        if x != word_to_check[i:i + window_size]:
            word_to_check = word_to_check.replace(word_to_check[i:i+window_size], x)
    return word_to_check


#for line in file:
line = "oneightwoneightwo"
line = check_word(line)
print(line)
nums = [i for i in line if i.isdigit()]
line_value = int(nums[0] + nums[-1])
cum_sum += line_value
print(cum_sum)

crude ledgeBOT Dec 8, 2023, 2:16 PM

#

@gray timber :white_check_mark: Your 3.12 eval job has completed with return code 0.

gray timber Dec 8, 2023, 2:16 PM

#

ah wait

#

12 is wrong for oneightwoneightwo

#

1igh2n8wo
18 right?

mortal siren Dec 8, 2023, 2:17 PM

#

no it should be 12

ancient cedar Dec 8, 2023, 2:17 PM

#

gray timber 12 is wrong for oneightwoneightwo

nope, 12 is correct because you still include overlaps

gray timber Dec 8, 2023, 2:18 PM

#

so oneightwoneightwo is 18282?

ancient cedar Dec 8, 2023, 2:19 PM

#

you could see it that way, yes

gray timber Dec 8, 2023, 2:19 PM

#

ah ok thx

ancient cedar Dec 8, 2023, 2:19 PM

#

as long as you're taking the "one" and "two" from the start and end respectively it's fine

gray timber Dec 8, 2023, 2:35 PM

#

wouldnt it be the easiest way to just put all the overlaps alos in my dict like "oneight": "18", etc.?

ancient cedar Dec 8, 2023, 2:45 PM

#

one workaround several people found was ||replacing one -> o1e, two -> t2o, etc. so the overlaps are preserved||

ancient cedar Dec 8, 2023, 2:46 PM

#

gray timber wouldnt it be the easiest way to just put all the overlaps alos in my dict like ...

how about "oneightwo" then?

#

or "threeighthreeighthree..."

gray timber Dec 8, 2023, 2:50 PM

#

ancient cedar or "threeighthreeighthree..."

ah yeah i would have an infinite dict then if i want to check for each case

gray timber Dec 8, 2023, 2:53 PM

#

ancient cedar one workaround several people found was ||replacing `one` -> `o1e`, `two` -> `t2...

man i want to look at it but also dont xD

mortal siren Dec 8, 2023, 5:05 PM

#

gray timber man i want to look at it but also dont xD

looks like you aren't handling the overlaps properly in the code still

#

ninesixmlfjxhscninehqcdvxf8nzfivetwonehhd
an example from my input - two is overwriting the o from one so you don't pick it up

gray timber Dec 8, 2023, 6:40 PM

#

mortal siren `ninesixmlfjxhscninehqcdvxf8nzfivetwonehhd` an example from my input - two is ov...

yeah because i didnt think about the overlaps at first, but i managed to do it a while ago, but thx 🙂

mortal siren Dec 8, 2023, 6:40 PM

#

cool :)

frigid nacelle Dec 10, 2023, 6:02 AM

#

I'm building up a for loop and trying to go through all of this. it's a bit of a head scratcher. This is where I gotten to:

for a in puzzle_input:
    calibrated_1st_value = int(a[0][0].isdigit())
    calibrated_2nd_value = int(a[0][-1].isdigit())

calibrated_values = calibrated_1st_value + calibrated_2nd_value

print(calibrated_values)```

Sigh. I'm a bit lost. I managed to convert puzzle input into the list so I could build a for loop out of it

#

What I understand from the day 1 site (https://adventofcode.com/2023/day/1)>

I need to pull 1st and last numbers from a string that can be only be int.
Add just 1st and last numbers as 1 number
Then make a sum of all the calibration values.

What doesn't make sense to me is that If I look at this example:
For example:

1abc2
pqr3stu8vwx
a1b2c3d4e5f
treb7uchet

that last line, 7. According to this site, the calibration would be 77, not 77. Is it because that 7 is a sole remaining number that it can be somehow turn into 77?

scenic jay Dec 10, 2023, 6:12 AM

#

the 1st number in treb7uchet is 7
the last number in treb7uchet is 7
hence, the calibration is 77

#

doesn't matter if they're the same digits

frigid nacelle Dec 10, 2023, 6:17 AM

#

Okay, fair enough, that makes sense.

#

Thanks STickie. Now to figuire out the rest.

frigid nacelle Dec 10, 2023, 9:34 AM

#

I'm still working this out but is this output correct?

#

['186', '5713', '736', '716', '711', '1262', '9351', '342124', '998', '77', '75447']

#

Nvm, no. This output is incorrect. Back to work then.

drowsy locust Dec 11, 2023, 6:51 PM

#

thoughts on my solution?

import re
import time

def main() -> None:
    with open(r"D:\01 Libraries\Documents\Coding\Advent of Code\Advent2023\Day 1\puzzle.txt",
              "r", encoding="utf-8") as f:
        puzzle: str = f.read()

    puzzle: list = puzzle.split("\n")
    total: int = 0
    for line in puzzle:
        digits = re.findall(r"\d", line)
        if len(digits) == 0:
            pass
        else:
            temp_digit = digits[0] + digits[-1]
            total += int(temp_digit)
    print(total)

if __name__ == "__main__":
    start = time.perf_counter()
    main()
    end = time.perf_counter()
    print('\n', end - start)

mortal siren Dec 11, 2023, 9:36 PM

#

drowsy locust thoughts on my solution? ```py import re import time def main() -> None: wi...

looks decent for p1!

if len(digits) == 0:
    pass

was this because you are encountering empty strings in the split? if so consider
puzzle: list = puzzle.splitlines()
which will ditch the final newline by default

#

Also, nice type hinting but usually you wouldn't just typehint a list on its own - you'd indicate what was in that list. In this case it's a list of strings so you'd do list[str]

drowsy locust Dec 11, 2023, 9:37 PM

#

mortal siren looks decent for p1! ```py if len(digits) == 0: pass ``` was this because yo...

no, it was to skip lines with no digits

mortal siren Dec 11, 2023, 9:38 PM

#

drowsy locust no, it was to skip lines with no digits

ah - fair but you won't likely have any such lines in the input - after all the puzzle does not tell you how to handle such cases

#

not a big deal though

drowsy locust Dec 11, 2023, 9:38 PM

#

mortal siren ah - fair but you won't likely have any such lines in the input - after all the ...

there were

#

I was getting index errors until I put that exception 😅

mortal siren Dec 11, 2023, 9:38 PM

#

drowsy locust there were

really? I highly doubt that

#

again, are you sure it wasn't just an empty line?

#

left over from the last split?

drowsy locust Dec 11, 2023, 9:39 PM

#

let me check to be sure

#

yeah

#

apparently it was for the last line which is empty

#

the error I mean

mortal siren Dec 11, 2023, 9:40 PM

#

yup - so that's where splitlines() helps

drowsy locust Dec 11, 2023, 9:40 PM

#

I just assumed there were lines without digits lol

mortal siren Dec 11, 2023, 9:40 PM

#

ha fair :)

drowsy locust Dec 11, 2023, 9:40 PM

#

mortal siren Also, nice type hinting but usually you wouldn't just typehint a list on its own...

didn't know this, thanks

mortal siren Dec 11, 2023, 9:40 PM

#

finally you can use timeit to time the solution easier

#

import timeit
print(timeit.timeit(main, number=1))

drowsy locust Dec 11, 2023, 9:43 PM

#

mortal siren ```py import timeit print(timeit.timeit(main, number=1)) ```

wouldn't that execute main twice?

mortal siren Dec 11, 2023, 9:43 PM

#

no - as long as you didn't call main beforehand :)

drowsy locust Dec 11, 2023, 9:44 PM

#

so it would be like this?

if __name__ == "__main__":
    print(timeit.timeit(main, number=1))

mortal siren Dec 11, 2023, 9:45 PM

#

that's basically how I do it, yup

#

one more thing about typehinting as well - typically you don't need to typehint many individual vars

#

total: int = 0 - the type hint here probably isn't necessary because most tools will be able to infer that total is an int simply because you assigned 0 to it

drowsy locust Dec 11, 2023, 9:47 PM

#

alright

mortal siren Dec 11, 2023, 9:47 PM

#

even if you were to run mypy on strict mode, it wouldn't insist that you type hint that var

drowsy locust Dec 11, 2023, 9:48 PM

#

mortal siren ```py import timeit print(timeit.timeit(main, number=1)) ```

more on this, what's the difference between subtracting two time.perfcounter(), subtracting two datetime.datetime.now(), and using timeit.timeit()?

mortal siren Dec 11, 2023, 9:50 PM

#

drowsy locust more on this, what's the difference between subtracting two `time.perfcounter()`...

timeit.timeit uses time.perfcounter internally - it's more of a convenience module so that you don't have to write out the calculating code

#

as for datetime.now, I suspect it's just not as precise

drowsy locust Dec 11, 2023, 9:53 PM

#

I see

mortal siren Dec 11, 2023, 9:53 PM

#

anyhow - have a go at p2 now :)

#

oh - perhaps one other thing to mention - your regex solution is fine, and actually a good idea considering p2 coming up. But if you were only doing P1, jumping straight to regex probably is overkill for solving this

drowsy locust Dec 11, 2023, 9:56 PM

#

why is it overkill?

mortal siren Dec 11, 2023, 9:56 PM

#

digits = [char for char in line if char.isdigit()] works just as well and is more readable

drowsy locust Dec 11, 2023, 9:56 PM

#

ah okay

#

to be honest

#

regex is my go to solution to anything that requires matching stuff

mortal siren Dec 11, 2023, 9:57 PM

#

in the end, you do you - but I would suggest not to jump to regex immediately

#

it's a great tool, but used sparingly - when the occasion calls for it

drowsy locust Dec 11, 2023, 9:57 PM

#

just out of complexity sake or something else?

mortal siren Dec 11, 2023, 9:57 PM

#

I think readability

#

regex patterns aren't always the most readable

#

in general, favour pure Python unless the resulting pure Python would be more complicated to express than the regex

drowsy locust Dec 11, 2023, 10:00 PM

#

ah okay

mortal siren Dec 11, 2023, 10:01 PM

#

P2 is an example where regex becomes more viable as a solution

#

just because the resulting pure Python might be quite a bit less readable

drowsy locust Dec 11, 2023, 10:04 PM

#

I see

#

tbh, I have used regex as a primary tool for so long that I kind of can just read it lol

mortal siren Dec 11, 2023, 10:07 PM

#

again, you do you - after all, it's AOC

#

that's just my thoughts on it

drowsy locust Dec 11, 2023, 10:07 PM

#

oh I understand. was jut giving my thoughts as well

drowsy locust Dec 12, 2023, 12:51 AM

#

@mortal siren code for the second part:

import re
import timeit

c_digs = {
    'one': '1', 'two': '2', 'three': '3',
    'four': '4', 'five': '5', 'six': '6',
    'seven': '7', 'eight': '8', 'nine': '9'
}

def main() -> None:
    with open(r"D:\01 Libraries\Documents\Coding\Advent of Code\Advent2023\Day 1\puzzle.txt",
              "r", encoding="utf-8") as f:
        puzzle: str = f.read()

    puzzle: list[str] = puzzle.splitlines()
    total = 0
    for line in puzzle:
        digits: list[str] = re.findall(r"\d|one|two|three|four|five|six|seven|eight|nine", line)
        if digits[0].isdigit() and digits[-1].isdigit():
            total += int(digits[0] + digits[-1])
        elif digits[0].isdigit() and not digits[-1].isdigit():
            total += int(digits[0] + c_digs[digits[-1]])
        elif not digits[0].isdigit() and digits[-1].isdigit():
            total += int(c_digs[digits[0]] + digits[-1])
        elif not digits[0].isdigit() and not digits[-1].isdigit():
            total += int(c_digs[digits[0]] + c_digs[digits[-1]])
    print(total)

if __name__ == "__main__":
    print(timeit.timeit(main, number=1))

I'm seing it flow in the debugger, and it should work, but AoC is saying that my resut is too low

mortal siren Dec 12, 2023, 12:52 AM

#

Firstly - does your code work for test input?

#

(it looks like it should but do check that first)

drowsy locust Dec 12, 2023, 12:54 AM

#

test input? wdym?

#

it outputs a result if that's what you're asking

mortal siren Dec 12, 2023, 12:54 AM

#

so in the puzzle description they give you examples

#

and what result you should expect from the examples

drowsy locust Dec 12, 2023, 12:55 AM

#

right

mortal siren Dec 12, 2023, 12:55 AM

#

whenever you have issues, the first thing to do is run your code over the examples and check that you get the result they say

#

I actually go to the trouble of writing automated tests for all example inputs, but that's generally overkill :)

drowsy locust Dec 12, 2023, 12:55 AM

#

ah okay

mortal siren Dec 12, 2023, 12:57 AM

#

so check that out and report back whether it failed any or passed them all

drowsy locust Dec 12, 2023, 1:00 AM

#

so

#

apparently splitlines() is not ignoring the empty string

#

#

@mortal siren

#

also, yes

#

the code I wrote prints what the example says it's supposed to

mortal siren Dec 12, 2023, 1:05 AM

#

drowsy locust apparently splitlines() is not ignoring the empty string

How did you type in your input?

drowsy locust Dec 12, 2023, 1:05 AM

#

drowsy locust Dec 12, 2023, 1:06 AM

#

mortal siren How did you type in your input?

mortal siren Dec 12, 2023, 1:06 AM

#

So you have an extra newline at the beginning and extra space at the end

#

To get rid of the extra newline at the beginning, type \ right after the opening triple quotes

scenic jay Dec 12, 2023, 1:07 AM

#

or you could attach a .strip() to the end of the string

mortal siren Dec 12, 2023, 1:07 AM

#

To get rid of the spaces at the end, remove the spaces at the end :) To the left of the closing triple quote

mortal siren Dec 12, 2023, 1:08 AM

#

scenic jay or you could attach a ``.strip()`` to the end of the string

Yes but this way they learn the intricacies of triple quoted strings :)

scenic jay Dec 12, 2023, 1:08 AM

#

fair

mortal siren Dec 12, 2023, 1:08 AM

#

But yes you right they can strip() too ofc

#

Anyway this isn’t the main thing

#

The main thing is

drowsy locust Dec 12, 2023, 1:08 AM

#

gotcha

mortal siren Dec 12, 2023, 1:08 AM

#

Have you considered the oneight case?

drowsy locust Dec 12, 2023, 1:09 AM

#

what about it?

#

ah

#

i see

#

but my logic still captured it appropriately

#

so what's the issue?

mortal siren Dec 12, 2023, 1:12 AM

#

drowsy locust but my logic still captured it appropriately

Did it?

#

If you test oneight on its own, do you get 18?

drowsy locust Dec 12, 2023, 1:13 AM

#

no

#

drowsy locust Dec 12, 2023, 1:15 AM

#

drowsy locust

as you can see here, it captured everything exactly like the example

mortal siren Dec 12, 2023, 1:16 AM

#

drowsy locust as you can see here, it captured everything exactly like the example

Right but you aren’t capturing eight

#

You need to capture both one and eight

drowsy locust Dec 12, 2023, 1:17 AM

#

ah

mortal siren Dec 12, 2023, 1:17 AM

#

Or put another way, overlapping numbers count

drowsy locust Dec 12, 2023, 1:19 AM

#

interesting the puzzle not mentioning that

#

https://cdn.discordapp.com/emojis/571233358013661185.webp?size=48

mortal siren Dec 12, 2023, 1:20 AM

#

drowsy locust interesting the puzzle not mentioning that

I’m completely with you on that one

#

I did consider this case and thought that overlapping should not count because one word counts as a digit, so if you count that as a unit it shouldn’t be allowed to share letters with neighbouring words

#

But it counts

#

So there

drowsy locust Dec 12, 2023, 1:21 AM

#

bs lol

mortal siren Dec 12, 2023, 1:22 AM

#

I think this puzzle is the most BS of any in AOC, though others disagree

#

And the fact that it’s day 1 puzzle makes that even worse

drowsy locust Dec 12, 2023, 1:22 AM

#

btw

mortal siren Dec 12, 2023, 1:22 AM

#

Hardest, most BS day 1 puzzle

#

But there it is

scenic jay Dec 12, 2023, 1:23 AM

#

I disagree with you solely because day 8 exists

drowsy locust Dec 12, 2023, 1:23 AM

#

is that if/elif block the most elegant solution for the problem?

scenic jay Dec 12, 2023, 1:23 AM

#

but yeah this was kinda bad too

mortal siren Dec 12, 2023, 1:23 AM

#

scenic jay I disagree with you solely because day 8 exists

I am aware of the existence of day 8 and still think this is worse

#

But there you go

scenic jay Dec 12, 2023, 1:23 AM

#

drowsy locust is that if/elif block the most elegant solution for the problem?

elegant is subjective

drowsy locust Dec 12, 2023, 1:23 AM

#

is there a way to do it in less lines?

mortal siren Dec 12, 2023, 1:24 AM

#

drowsy locust is that if/elif block the most elegant solution for the problem?

There’s better ways to write it - but you should focus on getting your code working first

scenic jay Dec 12, 2023, 1:24 AM

#

there's almost always to do it in one line sure, but that usually isn't good looking nor well written so I wouldn't aim for it

mortal siren Dec 12, 2023, 1:24 AM

#

Idk I think using dict.get would be good looking

drowsy locust Dec 12, 2023, 1:25 AM

#

mortal siren There’s better ways to write it - but you should focus on getting your code work...

for example?

mortal siren Dec 12, 2023, 1:25 AM

#

dict.get(key, default)

#

If the key exists in the dictionary, retrieve it - else retrieve the default value

drowsy locust Dec 12, 2023, 1:26 AM

#

mortal siren So there

back to https://regex101.com I go lol

mortal siren Dec 12, 2023, 1:26 AM

#

drowsy locust back to https://regex101.com I go lol

That is one approach - build a better regex

#

There is actually a way to make it work with your existing regex pattern

#

Let me know if you want to know about that

drowsy locust Dec 12, 2023, 1:26 AM

#

nah, I'll figure it out

#

and yeah, I know

#

I have done this kind of thing before

mortal siren Dec 12, 2023, 1:27 AM

#

Alrighty!

drowsy locust Dec 12, 2023, 1:27 AM

#

last letter/first letter matching

#

just don't remember off the cuff, because it was literally only once lol

mortal siren Dec 12, 2023, 1:28 AM

#

That’s not actually what I would suggest

#

You can make it work with literally your pattern unchanged

#

But change the code around it

#

Or change the pattern and don’t change the code

#

Both are viable options

drowsy locust Dec 12, 2023, 1:29 AM

#

ah okay

mortal siren Dec 12, 2023, 1:32 AM

#

And yeah, the get. For example you can do c_digs.get(digits[0], digits[0])

#

Are you familiar with dict.get at all?

drowsy locust Dec 12, 2023, 1:35 AM

#

used once for a problem long time ago, didn't like and it didn't really fit my use, so no

#

yay

#

with reddit's spoilers I did it

#

I didn't know enough about regex operations to do this without spoilers

#

didn't even what lookahead or non-overlapping was

scenic jay Dec 12, 2023, 2:01 AM

#

the lookahead is basically just there to avoid consuming the string as you match it

drowsy locust Dec 12, 2023, 2:01 AM

#

I didn't know that was a thing!

#

well, I guest htat's the point of the event after all

#

foce you to learn new things

#

well onto to day 2

#

@mortal siren if you wanna talk about better ways to do it than that if/elif block, I'm all ears

analog ice Dec 16, 2023, 10:46 AM

#

spasm solution https://raw.githubusercontent.com/P403n1x87/aoc/main/2023/1/code.pya

mortal siren Dec 16, 2023, 11:55 AM

#

drowsy locust <@388062601281798144> if you wanna talk about better ways to do it than that if/...

I went to sleep by this time but can still discuss this if you'd like

drowsy locust Dec 16, 2023, 6:58 PM

#

mortal siren I went to sleep by this time but can still discuss this if you'd like

would love to. we can do it here, or if you don't mind, in DMs?

mortal siren Dec 16, 2023, 6:58 PM

#

can do it here, but let's do it later - I want to finish day 16 first

drowsy locust Dec 16, 2023, 7:00 PM

#

np

quasi vapor Jan 2, 2024, 5:38 PM

#

Part 2 completed in Python 3 without regex:
https://github.com/TerraMonster5/Advent-of-Code/blob/main/2023/Day 1/day1.py

old ember Jan 13, 2024, 2:42 PM

#

"oh this question is easy... there's a part two YOU SCOUNDRELS"

patent steeple Jan 16, 2024, 12:12 PM

#

with open("input1.txt") as f:
    data2 = f.read()
    data2 = data2.splitlines()


list_of_matches = [
    "1",
    "2",
    "3",
    "4",
    "5",
    "6",
    "7",
    "8",
    "9",
    "0",
    "one",
    "two",
    "three",
    "four",
    "five",
    "six",
    "seven",
    "eight",
    "nine",
]


def str_num_to_int(stringnumber):
    dct2 = {
        "0": 0,
        "1": 1,
        "2": 2,
        "3": 3,
        "4": 4,
        "5": 5,
        "6": 6,
        "7": 7,
        "8": 8,
        "9": 9,
        "10": 10,
        "one": 1,
        "two": 2,
        "three": 3,
        "four": 4,
        "five": 5,
        "six": 6,
        "seven": 7,
        "eight": 8,
        "nine": 9,
    }

    return dct2[stringnumber]


print(data2[0])


def bob(lst):
    func_lst = []
    for line in lst:
        dct1 = {}
        for word in list_of_matches:
            if word in line:
                index = line.index(word)
                dct1[index] = word
                print(dct1)
            else:
                pass
        final_number = str(str_num_to_int(dct1[min(dct1)])) + str(
            str_num_to_int(dct1[max(dct1)])
        )
        func_lst.append(int(final_number))
        print(int(final_number))
    test = sum(func_lst)
    print(test)


bob(data2)

Apparently my day 1 part 2 solution returns +10 on the correct number, does anyone know why its returning 54540 instead of 54530 (the solution to the dataset of day1 part 2). Thanks.

old prism Jan 16, 2024, 3:01 PM

#

patent steeple ```py with open("input1.txt") as f: data2 = f.read() data2 = data2.split...

one2eightwo654nine2 (expected answer 12) could be a helpful test case
answer: ||a match might appear multiple times||

patent steeple Jan 16, 2024, 3:23 PM

#

old prism `one2eightwo654nine2` (expected answer `12`) could be a helpful test case answer...

Thank you, mine returns 19 which is obviously wrong, I'll work on it 👍

old ember Jan 16, 2024, 11:47 PM

#

Is it cheating to use Perl for part 1?

#

# aoc day one part one
use v5.36;

open my $fh, '<', 'input.txt' or die;

my $total = 0;
while (my $line = <$fh>) {
    my ($x) = ($line =~ /\D*(\d)/);
    my ($y) = ((reverse $line) =~ /\D*(\d)/);
    $total += "$x$y";
}
say $total;```

verbal hinge Jan 17, 2024, 12:26 AM

#

old ember Is it cheating to use Perl for part 1?

Not in the least. Use whatever it takes that makes you feel good about it.

copper inlet Jan 27, 2024, 10:51 AM

#

I know this is pretty far past, but this is my first bit of code and I wanted to share it

check = 0
value = 0
#this function uses a for loop to iterate through a string and extract all numeric characters
def numextract(current):
    extracted = ""
    for char in current:
        if char.isnumeric():
            extracted += char
    return extracted
#this function removes all numbers besides the outer two
def numslim(num):
    final = 0
    snum = str(num)
    fir = snum[0:1]
    firI = int(fir)
    lisbac = snum[::-1]
    sec = lisbac[0:1]
    secI = int(sec)
    hold = firI * 10 + secI
    final = hold + final
    return final
data = input("Dataset: ")
datalist = data.splitlines()
length = len(datalist)
while length != check:
    current = datalist[check]
    num = numextract(current)
    digit = numslim(num)
    value = value + digit
    check = check + 1
print("Calibration Value", value)

mortal siren Jan 27, 2024, 12:31 PM

#

copper inlet I know this is pretty far past, but this is my first bit of code and I wanted to...

Hey there - congrats on doing Day 1 Part 1 :) Would you be interested in feedback on your code?

copper inlet Jan 27, 2024, 12:34 PM

#

Not right now! Thank you for asking though. I know there are tools I'm not aware of that I could have done much better if I had used them, but I used what I know, and I think that's enough to be proud of for now

mortal siren Jan 27, 2024, 2:08 PM

#

copper inlet Not right now! Thank you for asking though. I know there are tools I'm not aware...

Fair enough :) And yes, you should be proud of every completed problem :)

thin comet Jan 31, 2024, 5:12 PM

#

file = open("aoc1.txt", "r")
data = file.readlines()
number_strings = {"one": "1", "two": "2", "three": "3", "four": "4", "five": "5", "six": "6", "seven": "7",
                  "eight": "8", "nine": "9"}
calibration_sum = 0


def reverse_check(line):
    temp_str = ""
    for char in line:
        if line[0].isnumeric():
            line_numbers.append(line[0])
            break
        temp_str = line[0:3]
        temp_str = temp_str[::-1]
        if temp_str in number_strings:
            line_numbers.append(number_strings[temp_str[0:3]])
            break
        temp_str = line[0:4]
        temp_str = temp_str[::-1]
        if temp_str in number_strings:
            line_numbers.append(number_strings[temp_str[0:4]])
            break
        temp_str = line[0:5]
        temp_str = temp_str[::-1]
        if temp_str in number_strings:
            line_numbers.append(number_strings[temp_str[0:5]])
            break
        else:
            line = line[1:]


for line in data:
    line_numbers = []
    temp_string = ""
    running = True
    while running:
        if len(line_numbers) > 1:
            break
        while line[0].isnumeric():
            line_numbers.append(line[0])
            line = line[::-1]
            reverse_check(line)
        if line[0:3] in number_strings:
            line_numbers.append(number_strings[line[0:3]])
            line = line[::-1]
            reverse_check(line)
        elif line[0:4] in number_strings:
            line_numbers.append(number_strings[line[0:4]])
            line = line[::-1]
            reverse_check(line)
        elif line[0:5] in number_strings:
            line_numbers.append(number_strings[line[0:5]])
            line = line[::-1]
            reverse_check(line)
        else:
            line = line[1:]

    calibration_sum += int(line_numbers[0] + line_numbers[len(line_numbers) - 1])
file.close()
print(calibration_sum)```

#

my full solution, im sure it's ugly but it works

smoky escarp Feb 8, 2024, 6:05 PM

#

this is similar to the other solution you posted where there is a lot of repeated code. You might want to think about how you could write a couple of functions that take in a parameter.

smoky escarp Feb 8, 2024, 6:06 PM

#

thin comet ```py file = open("aoc1.txt", "r") data = file.readlines() number_strings = {"on...

^

#AoC 2023 | Day 1 | Solutions & Spoilers

py3.12 woo

LINES = ['1abc2', 'pqr3stu8vwx', 'a1b2c3d4e5f', 'treb7uchet']

1abc2 pqr3stu8vwx a1b2c3d4e5f treb7uchet

1abc2
pqr3stu8vwx
a1b2c3d4e5f
treb7uchet