Advent of Code 2017

2016 / 2017 / 2018 / 2019 / 2020 / [2021][]

Warning: Spoilers

Reflections and Benchmarks

I try to reflect on every day's puzzle, attempting to describe my thought processes and how my solutions all work. Benchmarks also included.

• Day 1 Reflections (code) (benchmarks)
• Day 2 Reflections (code) (benchmarks)
• Day 3 Reflections (code) (benchmarks)
• Day 4 Reflections (code) (benchmarks)
• Day 5 Reflections (code) (benchmarks)
• Day 6 Reflections (code) (benchmarks)
• Day 7 Reflections (code) (benchmarks)
• Day 8 Reflections (code) (benchmarks)
• Day 9 Reflections (code) (benchmarks) (stream)
• Day 10 Reflections (code) (benchmarks) (stream)
• Day 11 Reflections (code) (benchmarks)
• Day 12 Reflections (code) (benchmarks)
• Day 13 Reflections (code) (benchmarks)
• Day 14 Reflections (code) (benchmarks)
• Day 15 Reflections (code) (benchmarks)
• Day 16 Reflections (code) (benchmarks)
• Day 17 Reflections (code) (benchmarks)
• Day 18 Reflections (code) (benchmarks)
• Day 19 Reflections (code) (benchmarks)
• Day 20 Reflections (code) (benchmarks)
• Day 21 Reflections (code) (benchmarks)
• Day 22 Reflections (code) (benchmarks)
• Day 23 Reflections (code) (benchmarks)
• Day 24 Reflections (code) (benchmarks)
• Day 25 Reflections (code) (benchmarks)

Executable

Comes with test examples given in problems.

You can install using stack:

$ git clone https://github.com/mstksg/advent-of-code-2017
$ cd advent-of-code-2017
$ stack setup
$ stack install

The executable aoc2017 includes a testing and benchmark suite

$ aoc2017 --help
aoc2017 - Advent of Code 2017 challenge runner

Usage: aoc2017 DAY [PART] [-t|--tests] [-b|--bench]
  Run challenges from Advent of Code 2017

Available options:
  DAY                      Day of challenge (1 - 25), or "all"
  PART                     Challenge part (a, b, c, etc.)
  -t,--tests               Run sample tests
  -b,--bench               Run benchmarks
  -h,--help                Show this help text

$ aoc2017 5 b
>> Day 05b
>> [✓] 27720699

Benchmarking is implemented using criterion

$ aoc2017 2 --bench
>> Day 02a
benchmarking...
time                 729.1 μs   (695.0 μs .. 784.2 μs)
                     0.967 R²   (0.926 R² .. 0.995 R²)
mean                 740.4 μs   (711.9 μs .. 783.6 μs)
std dev              116.8 μs   (70.44 μs .. 172.8 μs)
variance introduced by outliers: 89% (severely inflated)

>> Day 02b
benchmarking...
time                 782.4 μs   (761.3 μs .. 812.9 μs)
                     0.983 R²   (0.966 R² .. 0.998 R²)
mean                 786.7 μs   (764.1 μs .. 849.4 μs)
std dev              110.8 μs   (42.44 μs .. 228.5 μs)
variance introduced by outliers: 86% (severely inflated)

Test suites run the example problems given in the puzzle description, and outputs are colorized in ANSI terminals.

$ aoc2017 1 --tests
[9] [!35732] $ aoc2017 1 --tests
>> Day 01a
[✓] (3)
[✓] (4)
[✓] (0)
[✓] (9)
[✓] Passed 4 out of 4 test(s)
[✓] 1097
>> Day 01b
[✓] (6)
[✓] (0)
[✓] (4)
[✓] (12)
[✓] (4)
[✓] Passed 5 out of 5 test(s)
[✓] 1188

This should only work if you're running aoc2017 in the project directory.

To run on actual inputs, the executable expects inputs to be found in the folder data/XX.txt in the directory you are running in. That is, the input for Day 7 will be expected at data/07.txt.

aoc2017 will download missing input files, but requires a session token. This can be provided in aoc2017-conf.yaml:

session:  [[ session token goes here ]]

You can "lock in" your current answers (telling the executable that those are the correct answers) by passing in --lock. This will lock in any final puzzle solutions encountered as the verified official answers. Later, if you edit or modify your solutions, they will be checked on the locked-in answers.

These are store in data/ans/XXpart.txt. That is, the target output for Day 7 (Part 2, b) will be expected at data/ans/07b.txt. You can also manually edit these files.