Distance from every row of x to every row of y

Measures the full cross product rather than walking the two arrays in lockstep, giving one list of length(y) distances per row of x.

Usage

ga_cross_distance(x, y, metric = "euclidean")

Arguments

x

a GeoArrow geometry array for metric = "euclidean", a point array for the others

y

a GeoArrow array matching x

metric

one of "euclidean", "haversine", "geodesic", "rhumb", or "vincenty"

Value

a list array with one element per row of x, each holding length(y) distances

Details

This is the shape ga_dist_euclidean_pairwise() cannot express, and it costs length(x) * length(y) distances to hold: ten thousand rows against ten thousand is a hundred million doubles, or eight hundred megabytes. Where the distances are only wanted to pick a nearest row or a threshold, ga_sparse_knn() and ga_sparse_dwithin() answer that against an R-tree without ever forming the product.

"euclidean" measures between geometries of any type. The spherical and ellipsoidal metrics take points only, because geo defines them between points alone. "vincenty" gives a null where the algorithm fails to converge.

A null row of x gives a null element. A null row of y gives a null in that position of every element.

See also

Other distance: ga_dist_euclidean_pairwise(), ga_dist_frechet_pairwise(), ga_dist_hausdorff_pairwise(), ga_dist_vincenty_pairwise()

Examples

x <- ga_xy(c(-78.6382, -80.8431), c(35.7796, 35.2271))
y <- ga_xy(c(-77.9447, -78.6382, -80.8431), c(34.2257, 35.7796, 35.2271))

# two elements of three distances each, in meters
ga_cross_distance(x, y, "haversine")
#> <nanoarrow_array list[2]>
#>  $ length    : int 2
#>  $ null_count: int 0
#>  $ offset    : int 0
#>  $ buffers   :List of 2
#>   ..$ :<nanoarrow_buffer validity<bool>[null] ``
#>   ..$ :<nanoarrow_buffer data_offset<int32>[3][12 b]> `0 3 6`
#>  $ children  :List of 1
#>   ..$ item:<nanoarrow_array double[6]>
#>   .. ..$ length    : int 6
#>   .. ..$ null_count: int 0
#>   .. ..$ offset    : int 0
#>   .. ..$ buffers   :List of 2
#>   .. .. ..$ :<nanoarrow_buffer validity<bool>[null] ``
#>   .. .. ..$ :<nanoarrow_buffer data<double>[6][48 b]> `183968 0 208827 28731...`
#>   .. ..$ dictionary: NULL
#>   .. ..$ children  : list()
#>  $ dictionary: NULL