Write a data frame to a vector dataset

The other direction of the same Arrow path: the data frame becomes one record batch, GDAL creates the fields from its schema, and OGR_L_WriteArrowBatch writes it.

Usage

write_vector(
  x,
  dsn,
  layer = NULL,
  driver = "GPKG",
  crs = NULL,
  geometry_type = "Unknown",
  geometry_column = NULL,
  fid_column = NULL,
  dataset_options = NULL,
  layer_options = NULL,
  write_options = NULL
)

Arguments

x

A data frame. A list column of raw vectors is the geometry.

dsn

Where to write it.

layer

The layer name to create. Defaults to the file's base name.

driver

The GDAL driver's short name.

crs

The coordinate reference system, in anything GDAL accepts: WKT, "EPSG:4326", a PROJ string. NULL for none.

geometry_type

The geometry type of the layer, by GDAL's name for it, for instance "Point" or "3D Polygon". "Unknown" lets the driver take whatever arrives.

geometry_column

The column holding WKB geometry. The default finds the one list column of raw vectors, which is what read_vector() returns.

fid_column

The column holding feature ids, which is a property of the layer rather than a field of it. The default uses a column called fid if there is one.

dataset_options, layer_options, write_options

Character vectors of KEY=VALUE options for dataset creation, layer creation and the write.

Value

dsn, invisibly.

Examples

places <- read_vector(system.file("extdata/test.gpkg", package = "GDAL7"))
path <- tempfile(fileext = ".gpkg")
write_vector(places, path, layer = "places", crs = "EPSG:4326")
identical(nrow(read_vector(path)), nrow(places))
#> [1] TRUE
unlink(path)