Solutions for Advent of Code
Anders Englöf Ytterström
d7d5312786
Turns out the initial trial of using Manhattan distance was required to solve part 2, since the BFS implementation is impossible to scale up from 2 robots to 25. Recursion and memoization makes the execution time acceptable. Line 53 makes all the difference here. The order of `<|v|^|>` matters when constructing a sequence. Many hours was spent trying to find the correct priority. For the example input, especially 456A and 379A were volatile. |
||
|---|---|---|
| 2015-elixir | ||
| 2015-elixir-2 | ||
| 2015-python | ||
| 2016-python | ||
| 2016-python2 | ||
| 2017-python | ||
| 2018-elixir | ||
| 2018-python | ||
| 2019-elixir | ||
| 2020-elixir | ||
| 2020-python | ||
| 2021-elixir | ||
| 2021-python | ||
| 2022-elixir | ||
| 2023-python | ||
| 2024-python | ||
| leaderboard | ||
| .gitignore | ||
| fetch-calendar | ||
| fetch-input | ||
| README.md | ||
Advent of Code 2015–
Here all solved puzzles for Advent of Code are kept for reference. Most solutions are written in Python, but there are also solutions written in Elixir.
- Initial runs from 2016–2020 are squashed in version control, so the original thoughts/rants about the solution are forgotten for future generations.
- 2021 onwards, each solution will have a separate commit. This also includes revisiting old events.
Where are the puzzle input files?
For many years, this repository had all puzzle inputs under version control. They are not anymore, although the scripts asumed they are downloaded and stored locally in a subdirectory, which is ignored by version control.
The reasons:
- Neither you or me do have permission to reproduce or redistribute them, they are not released under public domain or MIT/BSD/GPL or any other license.
- The author behind advent of code have kindly asked us who attend Advent of Code to not release them, since publiced puzzle inputs cen be used to copy the site and steal the AoC creator's work.
Sources: