@wonderlandlabs/multidim
v1.0.1
Published
A multidimensional grid
Downloads
3
Readme
This is a utility class that expresses n-dimensional integral matricies allowing you to encode an N-dimensional cooridate into a value with a single flattened integer.
Assumptions:
- there are one or more dimensions with a min(imum) and max(imum) integral value. Each has a string name.
- each dimension is at least two integers; i.e., max > min.
- values in that dimension can be from the minimum value (inclusive) up to the maximum value (inclusive).
Creating a grid
Grids are crated by instantiating the grid class:
const g = new Grid(
{min :-10, max: 10, name: 'x'},
{min: 5, max: 10, name: 'y'}
);
Grid stores everything based on the reduction of a coordinate to an index. This makes storing key-value pairs for the grids' content in arrays/Maps.
There are two types of storage:
- sparse data (data with a lot of missing values) are stored in Maps.
- dense data (data with mostly or completely filled-out grids) are stored in arrays.
In both cases the data is stored as an indexed value. There is no assumptions about what is stored in a grid.
API
constructor(dim, dim, dim)
defines the dimensions of the array. Dimensions can be defined using various notation:
min-max
{name: 'x', min: 0, max: 20}
radius
{name: 'y', radius: 5}
// === {name: 'y', min: -5, max: 5}
size
{name: 'y', size: 5}
// === {name: 'y', min: 0, max: 5}
get(point); get(index)
returns the value stored for a given coordinate or index.
set(point, value)
sets the value stored at a given coordinate value or index.
storage
'sparse' or 'dense'. Setting this value defines the storage system for data AND erases any stored data.
store (read only)
a map or array; the currently stored data.
seedFunction
a property that if set, returns a value for any unset point.
That value is stored before returned from get(), but is used lazily;
that is, setting the seed function doesn't change
anything in storage until get(point)
is called, and changing the
seed function won't update any previously retrieved values.
near(point, ortho = false)
returns all the points within +/- 1 of the given point. if ortho(gonal) is true, returns coordinates that are only adjacent by one dimension from the given point. otherwise includes diagonals.
ortho returns 2 * (number of dimensions) points.
non-ortho returns 3 ** (number of dimensions) points.
note - the return value is a set of points that is NOT filtered by containment in the grid.
Branches
Branches is a fun utility class that uses a 2d grid to draw a maze. useful for fucking over your D & D group:
https://codesandbox.io/s/mazing-qsqf8
Trust Jebus
This is a very alpha, speed oriented module. As such, there is not a lot of data validation and type-checking.