Crate hilbert_2d

Rust functions for mapping between 1D and 2D space using the Hilbert curve, and its approximations.

When working with images and matrices, use the h2xy_discrete and xy2h_discrete functions:

use hilbert_2d::{h2xy_discrete, xy2h_discrete, Variant};
 
let (x, y) = h2xy_discrete(7, 2, Variant::Hilbert); // (1, 2)
let h = xy2h_discrete(2, 1, 2, Variant::Hilbert); // 13

When performing real-valued calculations, use the continuous functions instead:

use hilbert_2d::{h2xy_continuous_f64, Variant};
 
// Approaches the bottom-left corner
let (x1, y1) = h2xy_continuous_f64(0.0, Variant::Hilbert); 
// Approaches the bottom-right corner
let (x2, y2) = h2xy_continuous_f64(1.0, Variant::Hilbert); 

Some of the pattern variants of the Hilbert curve have also been implemented:

use hilbert_2d::{h2xy_continuous_f64, Variant};
 
// In the Liu L1 variant, both ends of the curve approach the center of the square
let (x1, y1) = h2xy_continuous_f64(0.0, Variant::Liu1); // (~0.5, ~0.5)
let (x2, y2) = h2xy_continuous_f64(1.0, Variant::Liu1); // (~0.5, ~0.5)

Enums

Variant

Indicates the pattern variant of the Hilbert curve to be constructed.

Functions

h2xy_continuous_f32	Maps from a 1D value to an approximate 2D coordinate, using the closest approachable limit of the Hilbert curve. Recommended for real-valued calculations.
h2xy_continuous_f64	Maps from a 1D value to an approximate 2D coordinate, using the closest approachable limit of the Hilbert curve. Recommended for real-valued calculations.
h2xy_discrete	Maps from a 1D index to a 2D coordinate, using a discrete approximation of the Hilbert curve. Recommended for images and matrices.
xy2h_continuous_f32	Maps from a 2D coordinate to an approximate 1D value, using the closest approachable limit of the Hilbert curve. Recommended for real-valued calculations.
xy2h_continuous_f64	Maps from a 2D coordinate to an approximate 1D value, using the closest approachable limit of the Hilbert curve. Recommended for real-valued calculations.
xy2h_discrete	Maps from a 2D coordinate to an 1D index, using a discrete approximation of the Hilbert curve. Recommended for images and matrices.