How to read a geojson file containing feature collections to leaflet-shiny directly

Question

My question is how to read a geojson file containing feature collections to leaflet-shiny. I have seen joe's github https://github.com/jcheng5/leaflet-shiny/blob/master/inst/examples/geojson/server.R but he did not use an external dataset but created the geojson manually. i am confused whether

1. Is that possible to read geojson file to leaflet-shiny directly?
2. If not, what does the structure of feature collections look like in shiny (in Joe's post it is multi-polygon) and how to create that in a faster and easier way?

Answer 1

You're probably looking to be able to manipulate the GeoJSON file directly in R-Shiny and R as opposed to reading a static file.

As previously mentioned, you can feed a string containing the GeoJSON to leaflet-shiny such as this GeoJSON FeatureCollection:

{
  "type": "FeatureCollection",
  "features": [
{
    "type": "Feature",
    "properties": {"party": "Republican"},
    "id": "North Dakota",
    "geometry": {
        "type": "Polygon",
        "coordinates": [[
            [-104.05, 48.99],
            [-97.22,  48.98],
            [-96.58,  45.94],
            [-104.03, 45.94],
            [-104.05, 48.99]
        ]]
    }
},
{
    "type": "Feature",
    "properties": {"party": "Democrat"},
    "id": "Colorado",
    "geometry": {
        "type": "Polygon",
        "coordinates": [[
            [-109.05, 41.00],
            [-102.06, 40.99],
            [-102.03, 36.99],
            [-109.04, 36.99],
            [-109.05, 41.00]
        ]]
    }
}
  ]
}

Then you can use RJSONIO::fromJSON to read this object in the format provided in the example and manipulate it in R such as this (Note: it appears that you have to add styles after reading the GeoJSON file as opposed to reading a GeoJSON FeatureCollection file that already has styles):

geojson <- RJSONIO::fromJSON(fileLocation)
geojson[[2]][[1]]$properties$style <- list(color = "red",fillColor = "red")
geojson[[2]][[2]]$properties$style <- list(color = "blue",fillColor = "blue")
geojson$style <- list(weight = 5,stroke = "true",fill = "true",opacity = 1,fillOpacity = 0.4)

This will give you the same R object if you had just entered this:

geojson <- list(
  type = "FeatureCollection",
  features = list(         
        list(
          type = "Feature",
          geometry = list(type = "MultiPolygon",
                          coordinates = list(
                            list(
                              list(
                                c(-109.05, 41.00),
                                c(-102.06, 40.99),
                                c(-102.03, 36.99),
                                c(-109.04, 36.99),
                                c(-109.05, 41.00)
                              )
                            )
                          )
          ),
          properties = list(
            party = "Democrat",
            style = list(
              fillColor = "blue",
              color = "blue"
            )
          ),
          id = "Colorado"
        ),
        list(
          type = "Feature",
          geometry = list(type = "MultiPolygon",
                          coordinates = list(
                            list(
                              list(  
                                c(-104.05, 48.99),
                                c(-97.22,  48.98),
                                c(-96.58,  45.94),
                                c(-104.03, 45.94),
                                c(-104.05, 48.99)
                              )
                            )
                          )
          ),
          properties = list(
            party = "Republican",
            style = list(
              fillColor = "red",
              color = "red"
            )
          ),
          id = "North Dakota"
        )
    ),
  style = list(
          weight = 5,
          stroke = "true",
          fill = "true",
          fillOpacity = 0.4
          opacity = 1
        ))

Answer 2

For question 1, refering to the API doc, you can provide instead directly a geojson string instead of a list like in the example.

The provided example also mentionned it with

You can also use a GeoJSON string value instead of a structured GeoJSON object like this one

You can use this recipe to read your GeoJSON file

For question 2, as question 1 is Ok, no issue.