Struct HeightGrid 

pub struct HeightGrid {
    pub origin_x: f64,
    pub origin_y: f64,
    pub dx: f64,
    pub dy: f64,
    pub ncols: usize,
    pub nrows: usize,
    pub heights: Vec<f64>,
}

Terrain heights on an axis-aligned, uniform grid in metres. Row-major, south-to-north: heights[i * ncols + j] is column j (east) of row i, where row 0 is the southernmost row and i increases northward, matching the frame’s +z = north convention shared by monitors and local_en_m. origin_y is the south edge; sampling height_at moves north as z increases. Position y is up; the grid lies in the x–z plane.

Fields

origin_x: f64

Frame-x of the western (first) column.

origin_y: f64

Frame-z of the southern (first) row.

dx: f64

Column spacing, metres.

dy: f64

Row spacing, metres.

ncols: usize

Number of columns (east).

nrows: usize

Number of rows (south→north).

heights: Vec<f64>

Row-major height samples, ncols * nrows, south→north.

Implementations

impl HeightGrid

pub fn new( origin_x: f64, origin_y: f64, dx: f64, dy: f64, ncols: usize, nrows: usize, heights: Vec<f64>, ) -> Result<Self>

Build and validate a grid.

pub fn height_at(&self, x: f64, z: f64) -> f64

Terrain height at a horizontal position, bilinear between samples; clamps at the grid edges.

pub fn from_ascii(text: &str) -> Result<Self>

A grid from an ASCII ESRI .asc raster (projected, metres): header ncols/nrows/xllcorner/yllcorner/cellsize, then row-major heights (north-to-south, west-to-east, i.e. “English reading order”: the first row is the north edge). Rows are reversed into the grid’s south-to-north storage so +z points north.

pub fn from_south_up( origin_x: f64, origin_south_y: f64, dx: f64, dy: f64, ncols: usize, nrows: usize, heights: Vec<f64>, ) -> Result<Self>

A grid from raw metres data already in the grid’s south-to-north row-major order (row 0 = south), with the origin at the south-west sample corner.

pub fn downsample(&self, target_dx: f64) -> Result<HeightGrid>

Coarsen the grid to roughly target_dx metre spacing, averaging each covered cell (a cheap supersampling → mean). The origin and extents are preserved; dx/dy become the downsampled spacing. Terrain wind needs no more than ~30 m cells, so a 1 m LiDAR DEM should almost always be downsized before adjustment.

Trait Implementations

impl Clone for HeightGrid

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for HeightGrid

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.