advent-of-code-jq VS aoc

Advent of Code (https://adventofcode.com/)

It's not a programming course per-se, but it's a great resource to master the skill of coding and problem solving.

It's just one part though, it won't teach you anything about architecturing a bigger system.

Hanukkah of Data is a series of data-themed puzzles, where you solve puzzles to move your way through a holiday-themed story using a fictional dataset. I think of it as "Advent of Code meets SQL Murder Mystery".

I've been making a CLI for advent of code ( https://adventofcode.com/ ) this week: https://github.com/VitamintK/wang-aoc-cli

It's been satisfying!

December started 12 days ago, and for my first year I decided to try the Advent of Code 2023, which is basically 1 programming problem everyday and they get harder and harder each day. I started HARD, I ate problems, day by day, until... day 10; things started getting pretty hard and couldn't do - I think - pretty average difficulty problems.

If you want some recommendations on how to learn it best, I really suggest jsut using it and googling how to do the things you dont know how to do as you work. Right now, the advent of code is happening and its one of the best ways to practice and learn: https://adventofcode.com/ Another thing you might want to try is Project Euler: https://projecteuler.net

Apart from the books, I went through Google's Comprehensive Rust course and dabbled with Advent of Code.

Advent of Code

Well, if you want to subject yourself to something as bad as a gamification process, I guess there's Advent of Code running these days: https://adventofcode.com/

So like many of you know, Advent of Code is happening right now, and I couldn't resist a dopamine hit we get just from writing pure code but with a twist: writing it in fully immutable and functional idiomatic Kotlin style. I heard someone on Reddit is trying the same so I probably not alone. So please roast my code

[LANGUAGE: jq] github

Personally I just stick to raw arrays for most days, usually with static storage (so pre-allocated). Keeps things simple! Here's my repo.

- C -

My original solution checked every character against every other character, which performed perfectly fine even for 14 chars, but it felt wrong.

- C -

First a straightforward solution with a double buffer, later realised you could do it in-place by just holding onto one value per row/column.

First wrote a short AWK program to convert the input to C and play with it while running a hopeless brute force attempt.

An array of int pairs which is deduplicated each iteration - a suitable approximation of a set. Code here.

- C -

Interesting though that we ended up with a very similar approach from different angles - it seems like you thought about the problem and came up with how to identify the digits based on segments, whereas I chose to look at the data detached from any meaning and found that intersection counts yielded a unique sorting key - but that's in fact the very same thing.

Nasty C# with Linq