This library will contain standard path finding algorithms and return the resulting graph object. You can search for paths in a graph based or grid based structure.
Table of contents generated with markdown-toc
Currently supported:
Download the crate: https://crates.io/crates/path-finding
At the moment, we have following major concepts:
You only need to pass edges to the graph. The nodes are generated automatically. Each pathfinding method will accept a graph, and return a graph that only contains the edges and nodes of the result.
Alternatively, you can also create a graph if you provide an adjacency matrix. Edges and nodes will be generated automatically.
If you want to use the A* path-finding algorithm, please make sure to provide positional information for each node.
Additionally, we work on a support for each of those algorithms based on a 2D grid based structure.
rust
pub fn your_function() {
graph::Edge::from(
0 /* edge index */,
0 /* source node */,
1 /* destination node */,
0.1, /* weight */
);
}
rust
pub fn your_function() {
graph::Graph::from(Vec::from([edge1, edge2]));
}
```rust pub fn your_function() { let mut matrix: &[&[f32]] = &[ &[0.0, 4.0, 0.0, 0.0, 0.0, 0.0, 0.0, 8.0, 0.0], &[4.0, 0.0, 8.0, 0.0, 0.0, 0.0, 0.0, 11.0, 0.0], &[0.0, 8.0, 0.0, 7.0, 0.0, 4.0, 0.0, 0.0, 2.0], &[0.0, 0.0, 7.0, 0.0, 9.0, 14.0, 0.0, 0.0, 0.0], &[0.0, 0.0, 0.0, 9.0, 0.0, 10.0, 0.0, 0.0, 0.0], &[0.0, 0.0, 4.0, 14.0, 10.0, 0.0, 2.0, 0.0, 0.0], &[0.0, 0.0, 0.0, 0.0, 0.0, 2.0, 0.0, 1.0, 6.0], &[8.0, 11.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 7.0], &[0.0, 0.0, 2.0, 0.0, 0.0, 0.0, 6.0, 7.0, 0.0] ];
graph::Graph::from_adjacency_matrix(matrix);
} ```
You may want to get some information or mutate the graph in some way. Therefore, the graph currently supports three functions for convenience operations or to provide data for a heuristic function.
rust
pub fn your_function() {
let edges: Vec<Edge> = graph.sorted_by_weight_asc(); // will return a vector with edges ascending by weight
}
rust
pub fn your_function() {
// provide a hashmap, mapping the node id to a position - used for a* pathfinding heuristics
graph.offer_positions(HashMap::from([(1, Position::from(0.1, 0.2, 0.3))]));
}
In some cases, such as 2D games, a grid based structure is advantageous. The grid can be created by passing a cost
matrix to the construction function ::from. At each entry of the matrix you provide an unsigned integer, indicating
the effort it takes
to move to this position from any neighbouring position. In the example below moving from position (0, 0)to position
(0, 1) will require an effort, or require a cost, of 2. Equivalently, moving from (2, 2) to (1, 1) will incur a
cost of 1. Based on this information, we want to provide a support for grid based structures in each path finding
algorithm.
rust
pub fn your_function() {
let grid = grid::Grid::from(&[
&[4.0, 2.0, 1.0],
&[2.0, 1.0, 0.0],
&[3.0, 4.0, 7.0]
]);
}
Check if a position is outside the grid. Pass a coordinate to the function below.
rust
pub fn your_function() {
grid.outside((0, 1));
}
Check if a position is within the grid. Pass a coordinate to the function below.
rust
pub fn your_function() {
grid.within((0, 1));
}
Convert your coordinate to a node_id
rust
pub fn your_function() {
grid.node_id((0, 1));
}
Retrieve the cost required to move to a provided node id
rust
pub fn your_function() {
grid.cost(3);
}
rust
pub fn your_function() {
let mst_graph = graph::minimum_spanning(graph);
}
For graphs
rust
pub fn your_function() {
let dfs = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(DepthFirstSearch {}) /* used algorithm */
);
}
For grids
rust
pub fn your_function() {
let dfs = path::in_grid(
4 /* source */,
1 /* target */,
&grid,
Box::from(DepthFirstSearch {}) /* used algorithm */
);
}
For graphs
rust
pub fn your_function() {
let bfs = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(BreadthFirstSearch {}) /* used algorithm */
);
}
For grids
rust
pub fn your_function() {
let dfs = path::in_grid(
4 /* source */,
1 /* target */,
&grid,
Box::from(BreadthFirstSearch {}) /* used algorithm */
);
}
rust
pub fn your_function() {
let bi_bfs = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(BiBreadthFirstSearch {}) /* used algorithm */
);
}
rust
pub fn your_function() {
let dijkstra = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(Dijkstra {}) /* used algorithm */
);
}
You can use the A* path-finding algorithm by providing either an existing heuristic function as shown below. Or you provide your own heuristic function. In case you use an existing heuristic function, make sure to provide the positional information for the nodes.
rust
pub fn your_function_with_euclidean_distance() {
let a_star = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(AStar { heuristic: Box::from(euclidean_distance) }), /* used algorithm + euclidean distance heuristic function */
);
}
rust
pub fn your_function_with_manhattan_distance() {
let a_star = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(AStar { heuristic: Box::from(manhattan_distance) }), /* used algorithm + manhattan distance heuristic function */
);
}
Similar to the A* path-finding algorithm, you can provide either an existing heuristic function as shown in the previous section. Or you provide your own heuristic function. In case you use an existing heuristic function, make sure to provide the positional information for the nodes.
In addition to the functionality of A, this algorithm will require you to pass the graph to the Hierarchical A instance. The reason is simple, the algorithm will divide the graph into segments on creation and cache information required in the subsequent path-finding process.
rust
pub fn your_function_with_euclidean_distance() {
let a_star = path::in_graph(
4 /* source */,
1 /* target */,
&graph,
Box::from(HierarchicalAStar { heuristic, graph }), /* used algorithm and graph */
);
}