`sp::over()` equivalent in `terra`

Is there an equivalent of sp::over() in package terra? To get a data frame showing which geometries of a SpatVector overlay which geometries of another SpatVector -- like this, but using only terra:

# get a polygons map:
library(terra)
lux <- vect(system.file("ex/lux.shp", package="terra"))
plot(lux)
text(lux, lux$NAME_2)

# get points that overlay some of those polygons:
pts <- vect(cbind(c(5.8, 6, 6.2), c(49.85, 49.5, 49.6)), crs = crs(lux))
plot(pts, col = "blue", add = TRUE)

# find which points overlay which polygons:
library(sp); library(raster)
over(as(pts, "Spatial"), as(lux, "Spatial"))

#   ID_1     NAME_1 ID_2           NAME_2 AREA    POP
# 1    1   Diekirch    3          Redange  259  18664
# 2    3 Luxembourg    9 Esch-sur-Alzette  251 176820
# 3    3 Luxembourg   10       Luxembourg  237 182607

Solution 1:

Your example data

library(terra)
lux <- vect(system.file("ex/lux.shp", package="terra"))
pts <- vect(cbind(c(5.8, 6, 6.2), c(49.85, 49.5, 49.6)), crs = crs(lux))

You can use extract (also in raster)

extract(lux, pts)[,-1]
#  ID_1     NAME_1 ID_2           NAME_2 AREA    POP
#1    1   Diekirch    3          Redange  259  18664
#2    3 Luxembourg    9 Esch-sur-Alzette  251 176820
#3    3 Luxembourg   10       Luxembourg  237 182607

Or you can do (a variation on the solution by lovalery)

i <- relate(pts, lux, "within") |> apply(2, any)
lux[i,] |> data.frame()
#  ID_1     NAME_1 ID_2           NAME_2 AREA    POP
#1    1   Diekirch    3          Redange  259  18664
#2    3 Luxembourg    9 Esch-sur-Alzette  251 176820
#3    3 Luxembourg   10       Luxembourg  237 182607

In version 1.5-13 (currently the development version that you can install like this install.packages('terra', repos='https://rspatial.r-universe.dev')) that is now simpler with

i <- is.related(lux, pts, "intersects")
lux[i,] |> data.frame()
#  ID_1     NAME_1 ID_2           NAME_2 AREA    POP
#1    1   Diekirch    3          Redange  259  18664
#2    3 Luxembourg    9 Esch-sur-Alzette  251 176820
#3    3 Luxembourg   10       Luxembourg  237 182607

Solution 2:

Please find below one possible solution using the terra library

Reprex

• Code

library(terra)

values(lux)[apply(terra::relate(lux, pts, "covers"), 2, function(x) which(x == TRUE)),]

• Output

#>    ID_1     NAME_1 ID_2           NAME_2 AREA    POP
#> 3     1   Diekirch    3          Redange  259  18664
#> 10    3 Luxembourg    9 Esch-sur-Alzette  251 176820
#> 11    3 Luxembourg   10       Luxembourg  237 182607

^{Created on 2022-01-17 by the reprex package (v2.0.1)}