Use strings of geoJSON with leaflet

Question

I have a dataframe where my first columns is geojson strings and the second is a metric for that location. The string in the first column renders properly if pasted into https://geojson.io/

I want to plot this with leaflet within R, as shown in the following link. https://rstudio.github.io/leaflet/json.html

Unfortunately, I don't know how to get my data into a format that will work with leaflet (seemingly an sp object).

Example row of data:

geojson <- '{"type": "Polygon", "coordinates": [[ [-104.05, 48.99], [-97.22, 48.98], [-96.58, 45.94], [-104.03, 45.94], [-104.05, 48.99] ]]}'
measure1 <- 10000
test_df <- data.frame(geojson, measure1)
test_df$geojson <- as.character(test_df$geojson)

Any other tips on best practices in a situation like this would also be appreciated.

Answer 1

Pretty sure leaflet requires that the geojson have a properties slot. You can do that with geojson pkg, e.g.,

library(leaflet)
library(geojson)
geojson <- '{"type": "Polygon", "coordinates": [[ [-104.05, 48.99], [-97.22, 48.98], [-96.58, 45.94], [-104.03, 45.94], [-104.05, 48.99] ]]}'
geojson <- geojson::properties_add(geojson, population = 10000)

You can of course add the properties slot manually manipulating the string, but we use jqr, a fast JSON parser that will make sure to do it right

measure1 <- 10000
df <- data.frame(geojson, measure1, stringsAsFactors = FALSE)

leaflet() %>% 
  addTiles() %>% 
  addGeoJSON(df$geojson) %>% 
  setView(-100, 47.6, 7)

Answer 2

You can use library(geojsonsf) to convert the geojson column to an sfc column (and from then create an sf object)

Data

## construct data (I've added an extra row)
geojson <- c('{"type": "Polygon", "coordinates": [[ [-104.05, 48.99], [-97.22, 48.98], [-96.58, 45.94], [-104.03, 45.94], [-104.05, 48.99] ]]}',
             '{"type": "Polygon", "coordinates": [[ [-103.05, 47.99], [-96.22, 47.98], [-93.58, 41.94], [-104.03, 45.94], [-103.05, 47.99] ]]}')

measure1 <- c(10000,20000)
test_df <- data.frame(geojson, measure1)
test_df$geojson <- as.character(test_df$geojson)

sfc column

## create an 'sfc' column
test_df$geojson <- geojsonsf::geojson_sfc(test_df$geojson)

sf object

Use library(sf) to make it into an sf object

library(sf)
sf <- sf::st_sf(test_df)
sf
# Simple feature collection with 2 features and 1 field
# geometry type:  POLYGON
# dimension:      XY
# bbox:           xmin: -104.05 ymin: 41.94 xmax: -93.58 ymax: 48.99
# epsg (SRID):    4326
# proj4string:    +proj=longlat +datum=WGS84 +no_defs
# measure1                        geojson
# 1    10000 POLYGON ((-104.05 48.99, -9...
# 2    20000 POLYGON ((-103.05 47.99, -9...

Plot

Here you can do whatever you want with this sf object.

library(leaflet)

leaflet() %>%
    addTiles() %>%
    addPolygons(data = sf)

library(googleway)

set_key("YOUR_GOOGLE_MAP_KEY")

google_map() %>%
    add_polygons(data = sf)