This crate provides traits for use in the bracket-pathfinding and bracket-geometry crates. It is part of the overall bracket-lib system.
Using a trait interface means that bracket-lib doesn't have to know or care about how you store your data, and can still provide useful geometry and path-finding functions. Defaults are provided, allowing you to get up and running quickly.
Everything is exported via bracket_algorithm_traits::prelude.
A truly minimal implementation (replace TestMap with your map structure):
rust
struct TestMap{};
impl BaseMap for TestMap {}
impl Algorithm2D for TestMap{
fn dimensions(&self) -> Point {
Point::new(2, 2)
}
}
This is sufficient to provide the following services:
in_bounds(Point): is a point within the dimensions of the map?point2d_to_index(Point) -> usize: provides an array index for a point, within the map dimensions. It assumes that your array has striding by column.index_to_point2d(usize) -> Point : provides an x/y coordinate for a given array index, assuming the same striding.If you don't like the default implementations, feel free to override them.
There is an equivalent Algorithm3D for 3D grid-based maps (substitute Point3D for Point).
The BaseMap trait helps you define the map. If you want to use path-finding, you need to implement the is_opaque function:
rust
impl BaseMap for MyMap {
fn is_opaque(&self, _idx: usize) -> bool {
false
}
}
To support path-finding, you need to implement two more functions:
```rust impl BaseMap for MyMap { fn getavailableexits(&self, idx: usize) -> SmallVec<[(usize, f32); 10]> { Vec::new() // Provide a list of exit indices (you can use point2dtoindex to generate them) for this // tile inside the array. // Do NOT include the current tile as an available exit. }
fn get_pathing_distance(&self, _idx1: usize, _idx2: usize) -> f32 {
1.0
// This should be a distance, using whatever heuristic you prefer.
}
```