Convert input data frame into a spatial data frame
Arguments
- x
Data frame with a feature name column.
- location
Feature names (tidy evaluation).
- feature_type
The registered map corresponding to values in
location
. If NA (the default), the type is guessed from the values inlocation
.- geom_name
Name for the new column to contain the geometry.
Examples
add_geometry(nc_type_example_2, county, feature_type = "sf.nc")
#> Simple feature collection with 200 features and 2 fields
#> Geometry type: MULTIPOLYGON
#> Dimension: XY
#> Bounding box: xmin: -82.74389 ymin: 35.14903 xmax: -76.69376 ymax: 36.24614
#> Geodetic CRS: NAD27
#> First 10 features:
#> county type geometry
#> 1 MARTIN A MULTIPOLYGON (((-77.17846 3...
#> 2 ALAMANCE B MULTIPOLYGON (((-79.24619 3...
#> 3 BERTIE A MULTIPOLYGON (((-76.78307 3...
#> 4 CHATHAM B MULTIPOLYGON (((-79.55536 3...
#> 5 CHATHAM B MULTIPOLYGON (((-79.55536 3...
#> 6 HENDERSON B MULTIPOLYGON (((-82.57003 3...
#> 7 GASTON B MULTIPOLYGON (((-81.32282 3...
#> 8 GASTON B MULTIPOLYGON (((-81.32282 3...
#> 9 LINCOLN B MULTIPOLYGON (((-80.95677 3...
#> 10 HENDERSON A MULTIPOLYGON (((-82.57003 3...