Convert many input types with spatial data to geojson specified as a json string

Usage

geojson_json(
  input,
  lat = NULL,
  lon = NULL,
  group = NULL,
  geometry = "point",
  type = "FeatureCollection",
  convert_wgs84 = FALSE,
  crs = NULL,
  precision = NULL,
  ...
)

Arguments

input: Input list, data.frame, spatial class, or sf class. Inputs can also be dplyr tbl_df class since it inherits from data.frame.
lat: (character) Latitude name. The default is NULL, and we attempt to guess.
lon: (character) Longitude name. The default is NULL, and we attempt to guess.
group: (character) A grouping variable to perform grouping for polygons - doesn't apply for points
geometry: (character) One of point (Default) or polygon.
type: (character) The type of collection. One of 'auto' (default for 'sf' objects), 'FeatureCollection' (default for everything else), or 'GeometryCollection'. "skip" skips the coercion with package geojson functions; skipping can save significant run time on larger geojson objects. Spatial objects can only accept "FeatureCollection" or "skip". "skip" is not available as an option for numeric, list, and data.frame classes
convert_wgs84: Should the input be converted to the standard CRS system for GeoJSON (https://tools.ietf.org/html/rfc7946) (geographic coordinate reference system, using the WGS84 datum, with longitude and latitude units of decimal degrees; EPSG: 4326). Default is FALSE though this may change in a future package version. This will only work for sf or Spatial objects with a CRS already defined. If one is not defined but you know what it is, you may define it in the crs argument below.
crs: The CRS of the input if it is not already defined. This can be an epsg code as a four or five digit integer or a valid proj4 string. This argument will be ignored if convert_wgs84 is FALSE or the object already has a CRS.
precision: (integer) desired number of decimal places for coordinates. Using fewer decimal places decreases object sizes (at the cost of precision). This changes the underlying precision stored in the data. options(digits = <some number>) changes the maximum number of digits displayed (to find out what yours is set at see getOption("digits")); the value of this parameter will change what's displayed in your console up to the value of getOption("digits"). See Precision section for more.
...: Further args passed on to internal functions. For Spatial* classes, it is passed through to sf::st_write(). For sf classes, data.frames, lists, numerics, and geo_lists, it is passed through to jsonlite::toJSON()

Value

An object of class geo_json (and json)

Details

This function creates a geojson structure as a json character string; it does not write a file - see geojson_write() for that

Note that all sp class objects will output as FeatureCollection objects, while other classes (numeric, list, data.frame) can be output as FeatureCollection or GeometryCollection objects. We're working on allowing GeometryCollection option for sp class objects.

Also note that with sp classes we do make a round-trip, using sf::st_write() to write GeoJSON to disk, then read it back in. This is fast and we don't have to think about it too much, but this disk round-trip is not ideal.

For sf classes (sf, sfc, sfg), the following conversions are made:

sfg: the appropriate geometry Point, LineString, Polygon, MultiPoint, MultiLineString, MultiPolygon, GeometryCollection
sfc: GeometryCollection, unless the sfc is length 1, then the geometry as above
sf: FeatureCollection

Precision

Precision is handled in different ways depending on the class.

The digits parameter of jsonlite::toJSON controls precision for classes numeric, list, data.frame, and geo_list.

For sp classes, precision is controlled by sf::st_write, being passed down through geojson_write(), then through internal function write_geojson(), then another internal function write_ogr_sf()

For sf classes, precision isn't quite working yet.

Examples

if (FALSE) {
# From a numeric vector of length 2, making a point type
geojson_json(c(-99.74134244, 32.451323223))
geojson_json(c(-99.74134244, 32.451323223))[[1]]
geojson_json(c(-99.74134244, 32.451323223), precision = 2)[[1]]
geojson_json(c(-99.74, 32.45), type = "GeometryCollection")

## polygon type
### this requires numeric class input, so inputting a list will dispatch
### on the list method
poly <- c(
  c(-114.345703125, 39.436192999314095),
  c(-114.345703125, 43.45291889355468),
  c(-106.61132812499999, 43.45291889355468),
  c(-106.61132812499999, 39.436192999314095),
  c(-114.345703125, 39.436192999314095)
)
geojson_json(poly, geometry = "polygon")

# Lists
## From a list of numeric vectors to a polygon
vecs <- list(
  c(100.0, 0.0), c(101.0, 0.0), c(101.0, 1.0), c(100.0, 1.0),
  c(100.0, 0.0)
)
geojson_json(vecs, geometry = "polygon")

## from a named list
mylist <- list(
  list(latitude = 30, longitude = 120, marker = "red"),
  list(latitude = 30, longitude = 130, marker = "blue")
)
geojson_json(mylist, lat = "latitude", lon = "longitude")

# From a data.frame to points
geojson_json(us_cities[1:2, ], lat = "lat", lon = "long")
geojson_json(us_cities[1:2, ],
  lat = "lat", lon = "long",
  type = "GeometryCollection"
)

# from data.frame to polygons
head(states)
## make list for input to e.g., rMaps
geojson_json(states[1:351, ],
  lat = "lat", lon = "long", geometry = "polygon",
  group = "group"
)

# from a geo_list
a <- geojson_list(us_cities[1:2, ], lat = "lat", lon = "long")
geojson_json(a)

# sp classes

## From SpatialPolygons class
library("sp")
poly1 <- Polygons(list(Polygon(cbind(
  c(-100, -90, -85, -100),
  c(40, 50, 45, 40)
))), "1")
poly2 <- Polygons(list(Polygon(cbind(
  c(-90, -80, -75, -90),
  c(30, 40, 35, 30)
))), "2")
sp_poly <- SpatialPolygons(list(poly1, poly2), 1:2)
geojson_json(sp_poly)

## data.frame to geojson
geojson_write(us_cities[1:2, ], lat = "lat", lon = "long") %>% as.json()

# From SpatialPoints class
x <- c(1, 2, 3, 4, 5)
y <- c(3, 2, 5, 1, 4)
s <- SpatialPoints(cbind(x, y))
geojson_json(s)

## From SpatialPointsDataFrame class
s <- SpatialPointsDataFrame(cbind(x, y), mtcars[1:5, ])
geojson_json(s)

## From SpatialLines class
library("sp")
c1 <- cbind(c(1, 2, 3), c(3, 2, 2))
c2 <- cbind(c1[, 1] + .05, c1[, 2] + .05)
c3 <- cbind(c(1, 2, 3), c(1, 1.5, 1))
L1 <- Line(c1)
L2 <- Line(c2)
L3 <- Line(c3)
Ls1 <- Lines(list(L1), ID = "a")
Ls2 <- Lines(list(L2, L3), ID = "b")
sl1 <- SpatialLines(list(Ls1))
sl12 <- SpatialLines(list(Ls1, Ls2))
geojson_json(sl1)
geojson_json(sl12)

## From SpatialLinesDataFrame class
dat <- data.frame(
  X = c("Blue", "Green"),
  Y = c("Train", "Plane"),
  Z = c("Road", "River"), row.names = c("a", "b")
)
sldf <- SpatialLinesDataFrame(sl12, dat)
geojson_json(sldf)
geojson_json(sldf)

## From SpatialGrid
x <- GridTopology(c(0, 0), c(1, 1), c(5, 5))
y <- SpatialGrid(x)
geojson_json(y)

## From SpatialGridDataFrame
sgdim <- c(3, 4)
sg <- SpatialGrid(GridTopology(rep(0, 2), rep(10, 2), sgdim))
sgdf <- SpatialGridDataFrame(sg, data.frame(val = 1:12))
geojson_json(sgdf)

# From SpatialPixels
library("sp")
pixels <- suppressWarnings(
  SpatialPixels(SpatialPoints(us_cities[c("long", "lat")]))
)
summary(pixels)
geojson_json(pixels)

# From SpatialPixelsDataFrame
library("sp")
pixelsdf <- suppressWarnings(
  SpatialPixelsDataFrame(
    points = canada_cities[c("long", "lat")],
    data = canada_cities
  )
)
geojson_json(pixelsdf)

# From sf classes:
if (require(sf)) {
  ## sfg (a single simple features geometry)
  p1 <- rbind(c(0, 0), c(1, 0), c(3, 2), c(2, 4), c(1, 4), c(0, 0))
  poly <- rbind(c(1, 1), c(1, 2), c(2, 2), c(1, 1))
  poly_sfg <- st_polygon(list(p1))
  geojson_json(poly_sfg)

  ## sfc (a collection of geometries)
  p1 <- rbind(c(0, 0), c(1, 0), c(3, 2), c(2, 4), c(1, 4), c(0, 0))
  p2 <- rbind(c(5, 5), c(5, 6), c(4, 5), c(5, 5))
  poly_sfc <- st_sfc(st_polygon(list(p1)), st_polygon(list(p2)))
  geojson_json(poly_sfc)

  ## sf (collection of geometries with attributes)
  p1 <- rbind(c(0, 0), c(1, 0), c(3, 2), c(2, 4), c(1, 4), c(0, 0))
  p2 <- rbind(c(5, 5), c(5, 6), c(4, 5), c(5, 5))
  poly_sfc <- st_sfc(st_polygon(list(p1)), st_polygon(list(p2)))
  poly_sf <- st_sf(foo = c("a", "b"), bar = 1:2, poly_sfc)
  geojson_json(poly_sf)
}

## Pretty print a json string
geojson_json(c(-99.74, 32.45))
geojson_json(c(-99.74, 32.45)) %>% pretty()

# skipping the pretty geojson class coercion with the geojson pkg
if (require(sf)) {
  library(sf)
  p1 <- rbind(c(0, 0), c(1, 0), c(3, 2), c(2, 4), c(1, 4), c(0, 0))
  p2 <- rbind(c(5, 5), c(5, 6), c(4, 5), c(5, 5))
  poly_sfc <- st_sfc(st_polygon(list(p1)), st_polygon(list(p2)))
  geojson_json(poly_sfc)
  geojson_json(poly_sfc, type = "skip")
}
}