Year 2018 Day 1

Author: maneatingape

README.md

@@ -68,6 +68,7 @@ Performance is reasonable even on older hardware, for example a 2011 MacBook Pro
 | [2021](#2021) | 10 |
 | [2020](#2020) | 286 |
 | [2019](#2019) | 22 |
+| [2018](#2018) | in progress |
 | [2017](#2017) | 515 |
 | [2016](#2016) | 663 |
 | [2015](#2015) | 87 |
@@ -232,6 +233,12 @@ Performance is reasonable even on older hardware, for example a 2011 MacBook Pro
 | 24 | [Planet of Discord](https://adventofcode.com/2019/day/24) | [Source](src/year2019/day24.rs) | 139 |
 | 25 | [Cryostasis](https://adventofcode.com/2019/day/25) | [Source](src/year2019/day25.rs) | 2721 |
 
+## 2018
+
+| Day | Problem | Solution | Benchmark (μs) |
+| --- | --- | --- | --: |
+| 1 | [Chronal Calibration](https://adventofcode.com/2018/day/1) | [Source](src/year2018/day01.rs) | 15 |
+
 ## 2017
 
 ![pie-2017]

benches/benchmark.rs

@@ -128,6 +128,10 @@ mod year2017 {
     benchmark!(year2017, day25);
 }
 
+mod year2018 {
+    benchmark!(year2018, day01);
+}
+
 mod year2019 {
     benchmark!(year2019, day01);
     benchmark!(year2019, day02);

src/lib.rs

@@ -110,6 +110,11 @@ pub mod year2017 {
     pub mod day25;
 }
 
+/// # Travel through time to restore the seasonal timeline.
+pub mod year2018 {
+    pub mod day01;
+}
+
 /// # Rescue Santa from deep space with a solar system adventure.
 pub mod year2019 {
     pub mod day01;

src/main.rs

@@ -24,6 +24,7 @@ fn main() {
         .chain(year2015())
         .chain(year2016())
         .chain(year2017())
+        .chain(year2018())
         .chain(year2019())
         .chain(year2020())
         .chain(year2021())
@@ -176,6 +177,10 @@ fn year2017() -> Vec<Solution> {
     ]
 }
 
+fn year2018() -> Vec<Solution> {
+    vec![solution!(year2018, day01)]
+}
+
 fn year2019() -> Vec<Solution> {
     vec![
         solution!(year2019, day01),

src/year2018/day01.rs

@@ -0,0 +1,74 @@
+//! # Chronal Calibration
+//!
+//! The simplest approach to part two is to store previously seen numbers in a `HashSet` then
+//! stop once a duplicate is found. However this approach requires scanning the input of ~1,000
+//! numbers multiple times, around 150 times for my input.
+//!
+//! A much faster `O(nlogn)` approach relies on the fact that each frequency increases by the same
+//! amount (the sum of all deltas) each time the list of numbers is processed. For example:
+//!
+//! ```none
+//!    Deltas: +1, -2, +3, +1 =>
+//!    0    1   -1    2
+//!    3    4    2    5
+//! ```
+//!
+//! Two frequencies that are a multiple of the sum will eventually repeat. First we group each
+//! frequencies by its remainder modulo the sum, using `rem_euclid` to handle negative frequencies
+//! correctly, Then we sort, first by the remainder to group frequencies that can repeat together,
+//! then by the frequency increasing in order to help find the smallest gap between similar
+//! frequencies, then lastly by index as this is needed in the next step.
+//!
+//! For the example this produces `[(0, 0, 0), (1, 1, 1), (2, -1, 2), (2, 2, 3)]`. Then we use
+//! a sliding windows of size two to compare each pair of adjacent canditates, considering only
+//! candidates with the same remainder. For each valid pair we then produce a tuple of
+//! `(frequency gap, index, frequency)`.
+//!
+//! Finally we sort the tuples in ascending order, first by smallest frequency gap, breaking any
+//! ties using the index to find frequencies that appear earlier in the list. The first tuple
+//! in the list gives the result, in the example this is `[(3, 2, 2)]`.
+use crate::util::parse::*;
+
+pub fn parse(input: &str) -> Vec<i32> {
+    input.iter_signed().collect()
+}
+
+pub fn part1(input: &[i32]) -> i32 {
+    input.iter().sum()
+}
+
+pub fn part2(input: &[i32]) -> i32 {
+    // The frequencies increase by this amount each pass through the list of deltas.
+    let total: i32 = input.iter().sum();
+
+    // Calculate tuples of `(frequency gap, index, frequency)` then sort to group frequencies that
+    // can collide together.
+    let mut frequency: i32 = 0;
+    let mut seen = Vec::with_capacity(input.len());
+
+    for n in input {
+        seen.push((frequency.rem_euclid(total), frequency, seen.len()));
+        frequency += n;
+    }
+
+    seen.sort_unstable();
+
+    // Compare each adjacent pair of tuples to find candidates, then sort by smallest gap first,
+    // tie breaking with index if needed.
+    let mut pairs = Vec::new();
+
+    for window in seen.windows(2) {
+        let (remainder0, freq0, index0) = window[0];
+        let (remainder1, freq1, _) = window[1];
+
+        if remainder0 == remainder1 {
+            pairs.push((freq1 - freq0, index0, freq1));
+        }
+    }
+
+    pairs.sort_unstable();
+
+    // Result is the frequency of the first tuple.
+    let (_, _, freq) = pairs[0];
+    freq
+}

tests/test.rs

@@ -112,6 +112,10 @@ mod year2017 {
     mod day25_test;
 }
 
+mod year2018 {
+    mod day01_test;
+}
+
 mod year2019 {
     mod day01_test;
     mod day02_test;

tests/year2018/day01_test.rs

@@ -0,0 +1,19 @@
+use aoc::year2018::day01::*;
+
+const EXAMPLE: &str = "\
++1
+-2
++3
++1";
+
+#[test]
+fn part1_test() {
+    let input = parse(EXAMPLE);
+    assert_eq!(part1(&input), 3);
+}
+
+#[test]
+fn part2_test() {
+    let input = parse(EXAMPLE);
+    assert_eq!(part2(&input), 2);
+}