Add a GeoJSON/sf source • toro

Preparing the data

A GeoJSON source is a format used to represent geographical features (as well as their non-geographic attributes). These sources can hold features like polygons, lines, points, circles, and more. In R a GeoJSON object will contain the geojson class.

An sf object is a simple feature data frame that contains a geometry column along with other columns containing a variety of attributes. In R an sf object will contain the sf class.

For this example, we use the quakes dataset that comes with base R. quakes is, by default, just a data frame. We use the sf package to turn it into a simple feature data frame.

library(sf)
#> Linking to GEOS 3.13.0, GDAL 3.8.5, PROJ 9.5.1; sf_use_s2() is TRUE
library(geojsonsf)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
library(toro)

data(quakes)
quakes_data <- quakes |>
  dplyr::mutate(hover = paste0("Magnitude: ", mag, "</br>Depth: ", depth)) |>
  sf::st_as_sf(coords = c("long", "lat"), crs = 4326)

At this point the data is still not GeoJSON data (it is an sf object), however, the add_source() function contains the ability to turn an sf object into a geojson object. This means that we can pass quakes_data to add_source() and it will be internally converted from sf to geojson.

For the purpose of this example, we will split quakes_data into two objects:

An sf object containing quakes with a magnitude below 6
A geojson object containing quakes with a magnitude of 6 or above

# Keep as a sf object
sf_quakes_data <- quakes_data |>
  dplyr::filter(mag < 6)
print(sf_quakes_data)
#> Simple feature collection with 995 features and 4 fields
#> Geometry type: POINT
#> Dimension:     XY
#> Bounding box:  xmin: 165.67 ymin: -38.59 xmax: 188.13 ymax: -10.72
#> Geodetic CRS:  WGS 84
#> First 10 features:
#>    depth mag stations                         hover              geometry
#> 1    562 4.8       41 Magnitude: 4.8</br>Depth: 562 POINT (181.62 -20.42)
#> 2    650 4.2       15 Magnitude: 4.2</br>Depth: 650 POINT (181.03 -20.62)
#> 3     42 5.4       43  Magnitude: 5.4</br>Depth: 42     POINT (184.1 -26)
#> 4    626 4.1       19 Magnitude: 4.1</br>Depth: 626 POINT (181.66 -17.97)
#> 5    649 4.0       11   Magnitude: 4</br>Depth: 649 POINT (181.96 -20.42)
#> 6    195 4.0       12   Magnitude: 4</br>Depth: 195 POINT (184.31 -19.68)
#> 7     82 4.8       43  Magnitude: 4.8</br>Depth: 82   POINT (166.1 -11.7)
#> 8    194 4.4       15 Magnitude: 4.4</br>Depth: 194 POINT (181.93 -28.11)
#> 9    211 4.7       35 Magnitude: 4.7</br>Depth: 211 POINT (181.74 -28.74)
#> 10   622 4.3       19 Magnitude: 4.3</br>Depth: 622 POINT (179.59 -17.47)

# Transform into a geojson object
geojson_quakes_data <- quakes_data |>
  dplyr::filter(mag >= 6) |>
  geojsonsf::sf_geojson()
print(geojson_quakes_data)
#> {"type":"FeatureCollection","features":[{"type":"Feature","properties":{"depth":139,"mag":6.1,"stations":94,"hover":"Magnitude: 6.1</br>Depth: 139"},"geometry":{"type":"Point","coordinates":[169.92,-20.7]}},{"type":"Feature","properties":{"depth":50,"mag":6.0,"stations":83,"hover":"Magnitude: 6</br>Depth: 50"},"geometry":{"type":"Point","coordinates":[165.96,-13.64]}},{"type":"Feature","properties":{"depth":127,"mag":6.4,"stations":122,"hover":"Magnitude: 6.4</br>Depth: 127"},"geometry":{"type":"Point","coordinates":[167.62,-15.56]}},{"type":"Feature","properties":{"depth":242,"mag":6.0,"stations":132,"hover":"Magnitude: 6</br>Depth: 242"},"geometry":{"type":"Point","coordinates":[167.02,-12.23]}},{"type":"Feature","properties":{"depth":165,"mag":6.0,"stations":119,"hover":"Magnitude: 6</br>Depth: 165"},"geometry":{"type":"Point","coordinates":[170.56,-21.59]}}]}
#> $...
#> NULL

Adding a GeoJSON/sf source

There are two ways to add a GeoJSON/sf source to your map:

The add_source() function

Using the add_source() function, you can add a GeoJSON source by passing an sf or geojson object directly to the data argument. sf objects will be automatically converted to GeoJSON format when added as a source. You can then use this source when adding layers to your map by specifying the source argument in the layer functions.

You can pass either a mapor mapProxy object to add_source().

map() |>
  set_bounds(bounds = quakes_data) |>
  add_source("sf_quakes_data", data = sf_quakes_data) |>
  add_source("geojson_quakes_data", data = geojson_quakes_data) |>
  add_circle_layer(
    id = "quakes_circles_sf",
    source = "sf_quakes_data",
    hover_column = "hover",
    paint = get_paint_options("circle", list(colour = "orange"))
  ) |>
    add_circle_layer(
    id = "quakes_circles_geojson",
    source = "geojson_quakes_data",
    hover_column = "hover",
    paint = get_paint_options("circle", list(colour = "red"))
  )

Adding the source directly in the layer function

Alternatively, you can add a GeoJSON/sf source directly within a layer function by passing a sf object or GeoJSON string to the source argument. In this case the source ID will be automatically generated by appending the layer ID to source-. For example, if you add a layer with id = "my_layer" and source is a sf object, the source will be added to the map with the ID "source-my_layer". This allows you to add the source and layer in a single step, without needing to call add_source() separately.

map() |>
  set_bounds(bounds = quakes_data) |>
  add_circle_layer(
    id = "quakes_circles_sf",
    source = sf_quakes_data,
    hover_column = "hover",
    paint = get_paint_options("circle", list(colour = "orange"))
  ) |>
    add_circle_layer(
    id = "quakes_circles_geojson",
    source = geojson_quakes_data,
    hover_column = "hover",
    paint = get_paint_options("circle", list(colour = "red"))
  )