Advent of Code

The Advent of Code is an online programming contest that takes place every year in December. It works like an advent calendar, except each day instead of a piece of chocolate or a sweet you get a new problem to solve.

I have taken part in this contest for the first time this year, after an email from a colleague that mentioned prizes for the winner of our private leaderboard. Even without any prize, if it is a challenge then I must accept it!

Warning: if you are still trying to complete the problems, you may find minor spoilers ahead, but nothing game-breaking.

Choosing my weapons

As I was initially not planning on taking part in the challenge, I did not have much time to decide which language or tools to use. I opted for solving the problems in C, because it is the language I worked with the most recently, but in hindsight it would have been more convenient to refresh my Python or C++ skills.

In the end using a more limited language did not make a big difference: the easy problems were still easy, some of the hard ones became a bit slower to type out. I also had some fun implementing some basic data structures, for example heaps for day 17.

The fun

Overall, I really enjoyed solving these problems! I liked the diverse set of techniques that were needed: graph algorithms, dynamic programming, computational geometry… a very nice selection of brain teasers!

I also liked the fact that the difficulty increased on average, but sometimes a hard problem was be followed by an easier one. This way, even if I found a problem particularly hard, I could still hope that the next one would be quicker to solve.

I also liked that other people, including friends, colleagues and my girlfriend, were taking on the challenge at the same time. I enjoyed explaining my solution or asking my friends to explain theirs. By the way, my friend Jared has some in-depth explanation of his solutions in his blog - check it out!

The ugly

There were a couple of problems that I disliked, and all of them for the same reason: the problem was not solvable without taking advantage of specific properties of the input data that were not made explicit in the problem’s statement.

As a Mathematician, I am never going to randomly assume that a generic graph has a specific structure, or that just throwing the lcm into my algorithm would make it work. So I was scratching my head for hours trying to solve a general problem that was very likely unsolvable, when I only had to solve a special case.

However, as a Reddit user pointed out, the input data is part of the problem statement. Analyzing it to figure out what algorithm may work is a skill. I guess I learnt something from this.

Solutions

If you are interested, you can find all my solutions on my git page and on Github. They are written in C99 without any external dependency other than the C standard library.

Apart from the harder problems, I have not commented my solutions much, but you can send me an email if you want some explanation!