experimental-functions: Experimental Functions

In rgeos: Interface to Geometry Engine - Open Source ('GEOS')

Experimental Functions

Description

Functions still under development using the GEOS STRtree structure to find intersecting object component envelopes (bounding boxes).

Usage

    gUnarySTRtreeQuery(obj)
    gBinarySTRtreeQuery(obj1, obj2)
    poly_findInBoxGEOS(spl, as_points=TRUE)

Arguments

obj, obj1, obj2	Objects inheriting from either SpatialPolygons or SpatialLines, obj2 may also inherit from SpatialPoints
spl	Object that inherits from the SpatialPolygons class
as_points	Logical value indicating if the polygon should be treated as points

Details

gUnarySTRtreeQuery and poly_findInBoxGEOS do the same thing, but poly_findInBoxGEOS uses the as_points argument to build the input envelopes from proper geometries. gUnarySTRtreeQuery and gBinarySTRtreeQuery build input envelopes by disregarding topology and reducing the coordinates to a multipoint representation. This permits the tree to be built and queried even when some geometries are invalid. gUnarySTRtreeQuery and poly_findInBoxGEOS return a list of length (n-1) of 1-based indices only for the “greater than i” indices. gBinarySTRtreeQuery returns a list of the length of obj2 with 1-based indices of obj1.

Author(s)

Roger Bivand & Colin Rundel

Examples

if (require(maptools)) {
xx <- readShapeSpatial(system.file("shapes/fylk-val-ll.shp",
 package="maptools")[1], proj4string=CRS("+proj=longlat +datum=WGS84"))
a0 <- gUnarySTRtreeQuery(xx)
a0
nc1 <- readShapePoly(system.file("shapes/sids.shp", package="maptools")[1],
 proj4string=CRS("+proj=longlat +datum=NAD27"))
a2 <- gUnarySTRtreeQuery(nc1)
#a3 <- poly_findInBoxGEOS(nc1)
#all.equal(a2, a3)
a2
pl <- slot(nc1, "polygons")[[4]]
a5 <- gUnarySTRtreeQuery(pl)
a5
SG <- Sobj_SpatialGrid(nc1, n=400)$SG
obj1 <- as(as(SG, "SpatialPixels"), "SpatialPolygons")
a4 <- gBinarySTRtreeQuery(nc1, obj1)
plot(nc1, col="orange", border="yellow")
plot(obj1, angle=sapply(a4, is.null)*45, density=20, lwd=0.5, add=TRUE)
set.seed(1)
pts <- spsample(nc1, n=10, type="random")
res <- gBinarySTRtreeQuery(nc1, pts)
}

rgeos documentation built on July 26, 2023, 5:42 p.m.