This README is the one indexed from the repository at its latest ref, not from the tag for this version.

Collections

Collections is a Crystal shard providing generic, dependency-free data structures that come up again and again in puzzles and algorithm work:

• BinaryHeapMin / BinaryHeapMax — binary heaps (with heapsort)
• PriorityQueue — pop values in priority order
• Counter — a multiset / tally, inspired by Python's collections.Counter
• DisjointSet — union-find with path compression and union by rank
• Graph — an undirected graph with clique detection
• WeightedGraph — a weighted graph with Dijkstra shortest paths
• Grid — a 2D grid with neighbours, flood fill, regions and BFS pathfinding

Installation

Add this to your application's shard.yml:

dependencies:
  collections:
    github: Lillevang/collections
    version: ~> 0.2.5

Run shards install

Usage

require "collections"

Heaps

heap = Collections::BinaryHeapMin(Int32).new
heap.add([10, 20, 5])
heap.sort           # => [5, 10, 20]  (non-destructive)
heap.extract_root!  # => 5

Collections::BinaryHeapMax(Int32).new.tap(&.add([10, 20, 5])).extract_root! # => 20

PriorityQueue

The value and the priority are independent types — the value need not be comparable.

pq = Collections::PriorityQueue(String, Int32).new
pq.push("low", 5)
pq.push("high", 1)
pq.pop # => "high"

Counter

Counts are stored as Int64, so large tallies do not overflow. Missing keys read as 0.

counter = Collections::Counter(Char).new("mississippi".chars)
counter['s']           # => 4
counter['z']           # => 0
counter.most_common(2) # => [{'i', 4}, {'s', 4}]

DisjointSet (union-find)

ds = Collections::DisjointSet(Int32).new
ds.union(1, 2)
ds.union(2, 3)
ds.connected?(1, 3) # => true
ds.count            # => 1  (number of disjoint sets)

Graph

graph = Collections::Graph(Int32).new
graph.add_edge(1, 2)
graph.add_edge(1, 3)
graph.neighbors(1).map(&.value) # => [2, 3]

WeightedGraph + Dijkstra

graph = Collections::WeightedGraph(String, Int32).new
graph.add_edge("a", "b", 1)
graph.add_edge("b", "c", 2)
graph.shortest_path("a", "c") # => {3, ["a", "b", "c"]}
graph.dijkstra("a")           # => {"a" => 0, "b" => 1, "c" => 3}

Grid

grid = Collections::Grid(Char).from_string("...\n.#.\n...")
grid.neighbors(0, 0)          # orthogonal, in-bounds, unblocked cells
grid.neighbors(1, 1, diagonal: true)

if result = grid.shortest_path({0, 0}, {2, 2})
  distance, path = result
  puts distance
  grid.print_grid(path)
end

Examples

Runnable, Advent-of-Code-flavoured programs live in examples/. Run any of them with crystal run:

crystal run examples/lanternfish.cr

Example	Shows
grid_pathfinding.cr	Parse a maze from text and BFS the shortest path
lanternfish.cr	Counter as a fast-growing Int64 tally
connected_components.cr	DisjointSet clustering
dijkstra_routes.cr	WeightedGraph + Dijkstra shortest paths
heaps_and_priority_queue.cr	Heapsort and a priority queue
graph_cliques.cr	Graph clique detection

Development

Write code, good code preferred!

Run tests with: crystal spec

Contributing

1. Fork it (https://github.com/Lillevang/collections/fork)
2. Create your feature branch (git checkout -b my-new-feature)
3. Commit your changes (git commit -am 'Add some feature')
4. Push to the branch (git push origin my-new-feature)
5. Create a new Pull Request

License

Released under the MIT License. See LICENSE for details.

Contributors

• Lillevang - creator and maintainer