Reads the geometries together with the feature properties. Properties are
untyped in GeoJSON, so read_geojson() infers each column's arrow type by
scanning every feature.
Value
a nanoarrow_array_stream of a single record batch, holding the
property columns followed by a geometry column.
Details
read_geojson() narrows the geometry column to the most specific type that
holds every feature, promoting a singular type to its multi form when it
finds both. It turns a feature with a null geometry into a null element
and ignores it when choosing the type:
| geometry types in the file | column type |
| all points | geoarrow.point |
| points and multipoints | geoarrow.multipoint |
| all polygons | geoarrow.polygon |
| polygons and multipolygons | geoarrow.multipolygon |
| more than one family, or any collection | geoarrow.geometry |
Narrowing matters in practice: geoarrow.geometry is a dense union that
geoarrow's R bindings cannot yet convert, so they return a genuinely mixed
collection as its raw storage rather than as geometries.
read_geojson() adds property columns in the order it first sees their
keys. It writes a null where a feature lacks a key. It widens types across
features:
| values seen for a key | column type |
| booleans | Boolean |
| integers | Int64 |
| integers and reals | Float64 |
| strings, or any mix that cannot unify | Utf8 |
only null | Utf8, all null |
read_geojson() keeps nested arrays and objects, which have no arrow
analogue, as their raw JSON text.
Per RFC 7946 the coordinate reference system is always OGC:CRS84, so
read_geojson() sets that on the geometry column without reading anything
from the file.
See also
Other io:
read_flatgeobuf(),
read_shapefile()
Examples
path <- tempfile(fileext = ".geojson")
fp <- system.file("shape/nc.shp", package = "sf")
sf::st_write(
sf::st_read(fp, quiet = TRUE),
path,
quiet = TRUE
)
# read into a nanoarrow array stream
res <- read_geojson(path)
res
#> <nanoarrow_array_stream struct<AREA: double, PERIMETER: double, CNTY_: double, CNTY_ID: double, NAME: string, FIPS: string, FIPSNO: double, CRESS_ID: int64, BIR74: double, SID74: double, NWBIR74: double, BIR79: double, SID79: double, NWBIR79: double, geometry: geoarrow.multipolygon{list<polygons: list<rings: list<vertices: struct<x: double, y: double>>>>}>>
#> $ get_schema:function ()
#> $ get_next :function (schema = x$get_schema(), validate = TRUE)
#> $ release :function ()
# convert to a df
df <- as.data.frame(res)
head(df)
#> AREA PERIMETER CNTY_ CNTY_ID NAME FIPS FIPSNO CRESS_ID BIR74 SID74
#> 1 0.114 1.442 1825 1825 Ashe 37009 37009 5 1091 1
#> 2 0.061 1.231 1827 1827 Alleghany 37005 37005 3 487 0
#> 3 0.143 1.630 1828 1828 Surry 37171 37171 86 3188 5
#> 4 0.070 2.968 1831 1831 Currituck 37053 37053 27 508 1
#> 5 0.153 2.206 1832 1832 Northampton 37131 37131 66 1421 9
#> 6 0.097 1.670 1833 1833 Hertford 37091 37091 46 1452 7
#> NWBIR74 BIR79 SID79 NWBIR79
#> 1 10 1364 0 19
#> 2 10 542 3 12
#> 3 208 3616 6 260
#> 4 123 830 2 145
#> 5 1066 1606 3 1197
#> 6 954 1838 5 1237
#> geometry
#> 1 <MULTIPOLYGON (((-81.4727554 36.2343559, -81.5408401 36.2725067, -81.56>
#> 2 <MULTIPOLYGON (((-81.2398911 36.3653641, -81.2406921 36.3794174, -81.26>
#> 3 <MULTIPOLYGON (((-80.4563446 36.2425575, -80.476387 36.2547264, -80.536>
#> 4 <MULTIPOLYGON (((-76.0089722 36.3195953, -76.0173492 36.3377304, -76.03>
#> 5 <MULTIPOLYGON (((-77.2176666 36.2409821, -77.2346115 36.2145996, -77.29>
#> 6 <MULTIPOLYGON (((-76.7450638 36.2339172, -76.98069 36.2302361, -76.9947>