The aim of this project is to provide a path-finding library that can be easily incorporated into web games. It may run on Node.js or the browser.
It comes along with an online demo to show how the algorithms execute. (The pathfinding speed is slowed down in the demo)
Note that this project only provides path-finding algorithms for 2D space. If you need to work in a 3D environment, then you may use @schteppe's fork.
There is new documentation being written for PathFinding.js. You can read it here. Note that this is in very early stages and far from complete so keep your eyes open for mistakes and don't hesitate to open a pull request in case you find one.
If you want to use it in Node.js, you may install it via npm
.
npm install pathfinding
Then, in your program:
var PF = require('pathfinding');
See the Basic Usage
section below for usage details.
If you have bower installed then you can install it with the following command:
bower install pathfinding
By default bower will install pathfinding under the bower_components folder, so to include it in your page do something like:
<script type="text/javascript" src="path/to/bower_components/pathfinding/pathfinding-browser.min.js"></script>
You can also grab a release from the Releases Page if you don't use bower.
To build a grid-map of width 5 and height 3:
var grid = new PF.Grid(5, 3);
By default, all the nodes in the grid will be able to be walked through.
To set whether a node at a given coordinate is walkable or not, use the setWalkableAt
method.
For example, to set the node at (0, 1) to be un-walkable, where 0 is the x coordinate (from left to right), and 1 is the y coordinate (from up to down):
grid.setWalkableAt(0, 1, false);
You may also pass in a matrix while instantiating the PF.Grid
class.
It will initiate all the nodes in the grid with the same walkability indicated by the matrix.
0 for walkable while 1 for blocked.
var matrix = [
[0, 0, 0, 1, 0],
[1, 0, 0, 0, 1],
[0, 0, 1, 0, 0],
];
var grid = new PF.Grid(matrix);
Currently there are 10 path-finders bundled in this library, namely:
-
AStarFinder
* BestFirstFinder
-
BreadthFirstFinder
* -
DijkstraFinder
* -
IDAStarFinder.js
* -
JumpPointFinder
* -
OrthogonalJumpPointFinder
* BiAStarFinder
BiBestFirstFinder
-
BiBreadthFirstFinder
* -
BiDijkstraFinder
*
The prefix Bi
for the last four finders in the above list stands for the bi-directional searching strategy.
Also, Note that only the finders with trailing asterisks are guaranteed to find the shortest path.
To build a path-finder, say, the AStarFinder
:
var finder = new PF.AStarFinder();
To find a path from (1, 2) to (4, 2), (Note: both the start point and end point should be walkable):
var path = finder.findPath(1, 2, 4, 2, grid);
path
will be an array of coordinates including both the start and end positions.
For the matrix
defined previously, the path
will be:
[ [ 1, 2 ], [ 1, 1 ], [ 2, 1 ], [ 3, 1 ], [ 3, 2 ], [ 4, 2 ] ]
Be aware that grid
will be modified in each path-finding, and will not be usable afterwards. If you want to use a single grid multiple times, create a clone for it before calling findPath
.
var gridBackup = grid.clone();
When instantiating path-finders, you may pass in additional parameters to indicate which specific strategies to use.
For all path-finders, you may indicate whether diagonal movement is allowed. The default value is false
, which means that the path can only go orthogonally.
In order to enable diagonal movement:
var finder = new PF.AStarFinder({
allowDiagonal: true
});
When diagonal movement is enabled, you might want to prevent the path from touching the corners of the occupied grid blocks. This is usually desirable if the objects using the path have physical width and can also move between the grid cells.
To enable the corner crossing prevention:
var finder = new PF.AStarFinder({
allowDiagonal: true,
dontCrossCorners: true
});
Note that dontCrossCorners
only makes sense when allowDiagonal
is also used. Currently all algorithms except JumpPointFinder
support this feature.
For AStarFinder
, BestFirstFinder
and all their Bi
relatives, you may indicate which heuristic function to use.
The predefined heuristics are PF.Heuristic.manhattan
(default), PF.Heuristic.chebyshev
, PF.Heuristic.euclidean
and PF.Heuristic.octile
.
To use the chebyshev heuristic:
var finder = new PF.AStarFinder({
heuristic: PF.Heuristic.chebyshev
});
To build a BestFirstFinder
with diagonal movement allowed and a custom heuristic function:
var finder = new PF.BestFirstFinder({
allowDiagonal: true,
heuristic: function(dx, dy) {
return Math.min(dx, dy);
}
});
To smoothen the path, you may use PF.Util.smoothenPath
. This routine will return
a new path with the original one unmodified.
var newPath = PF.Util.smoothenPath(grid, path);
Note that the new path will be compressed as well, i.e. if the original path is
[[0, 1], [0, 2], [0, 3], [0, 4]]
, then the new path will be [[0, 1], [0, 4]]
.
To just compress a path without smoothing it, you may use PF.Util.compressPath
.
var newPath = PF.Util.compressPath(path);
To expand the compressed path like [[0, 1], [0, 4]]
back to [[0, 1], [0, 2], [0, 3], [0, 4]]
,
you may use PF.Util.expandPath
.
var newPath = PF.Util.expandPath(path);
Layout:
.
|-- lib # browser distribution
|-- src # source code (algorithms only)
|-- test # test scripts
|-- utils # build scripts
|-- benchmark # benchmarks
`-- visual # visualization
Make sure you have node.js
installed, then use npm
to install the dependencies:
npm install -d
The build system uses gulp, so make sure you have it installed:
npm install -d -g gulp
To build the browser distribution:
gulp compile
To run the tests (algorithms only, not including the visualization) with mocha and should.js First install mocha:
npm install -d -g mocha
Then run the tests:
gulp test
To run the benchmarks:
gulp bench
Or if you are feeling lazy, the default gulp task does everything(except running the benchmarks):
gulp
© 2011-2012 Xueqiao Xu <xueqiaoxu@gmail.com>
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.