Interactive GeoLibre map

This page runs the same geolibre code used in RStudio, Quarto, R Markdown, and Shiny. The widgets below embed the hosted GeoLibre application, so viewing the live maps requires JavaScript and internet access.

library(geolibre)

Load a GeoLibre project

A complete .geolibre.json project can be loaded from a URL and passed to geolibre(). This example opens a 3D map of New York City buildings and subway lines. The project is fetched while this page is built, so the chunk reports the error rather than failing the build if the host is unreachable.

project_url <- "https://assets.geolibre.app/projects/nyc-buildings.geolibre.json"
nyc_buildings <- jsonlite::read_json(project_url, simplifyVector = FALSE)

geolibre(nyc_buildings, panels = "collapsed", height = 650)

A single point

Style overrides can be passed as named arguments or through a style list.

point <- list(
  type = "Feature",
  properties = list(name = "Washington, DC"),
  geometry = list(
    type = "Point",
    coordinates = c(-77.0369, 38.9072)
  )
)

geolibre(layout = "maponly", height = 600) |>
  add_geojson(point, name = "Washington, DC", fillColor = "#dc2626", circleRadius = 8) |>
  set_view(center = c(-77.0369, 38.9072), zoom = 10)

Points from a data frame

add_xy_data() reads longitude and latitude columns and keeps the remaining columns as feature properties, so they appear when a point is clicked. fit_bounds() frames the result.

cities <- data.frame(
  name = c("Washington", "New York", "Boston", "Philadelphia"),
  population = c(689545, 8336817, 654776, 1603797),
  longitude = c(-77.0369, -74.0060, -71.0589, -75.1652),
  latitude = c(38.9072, 40.7128, 42.3601, 39.9526)
)

geolibre(layout = "maponly", height = 600, basemap = "positron") |>
  add_circle_markers(cities, name = "Cities", radius = 9, fillColor = "#2563eb") |>
  fit_bounds(c(-78, 38, -70, 43))

A choropleth with a legend

add_choropleth() classifies a numeric column and colors it from a named ramp, computing the same graduated stops the application’s Style panel produces. color_ramp_names() lists the ramps, and interpolate_ramp_colors() samples one so a legend can match the map.

color_ramp_names()
#>  [1] "viridis"  "plasma"   "inferno"  "magma"    "cividis"  "turbo"   
#>  [7] "spectral" "blues"    "greens"   "oranges"  "reds"     "purples" 
#> [13] "terrain"  "rdylgn"   "rdylbu"   "rdbu"     "coolwarm" "jet"     
#> [19] "greys"    "gray"

counties <- list(
  type = "FeatureCollection",
  features = list(
    list(
      type = "Feature",
      properties = list(name = "Low", value = 10),
      geometry = list(type = "Point", coordinates = c(-77.5, 38.5))
    ),
    list(
      type = "Feature",
      properties = list(name = "Middle", value = 55),
      geometry = list(type = "Point", coordinates = c(-77.0, 39.0))
    ),
    list(
      type = "Feature",
      properties = list(name = "High", value = 100),
      geometry = list(type = "Point", coordinates = c(-76.5, 39.5))
    )
  )
)

breaks <- interpolate_ramp_colors("blues", 3)

geolibre(layout = "maponly", height = 600) |>
  add_choropleth(
    counties,
    column = "value",
    name = "Observations",
    colormap = "blues",
    class_count = 3,
    circleRadius = 14
  ) |>
  add_legend(
    "Observations",
    labels = c("10", "55", "100"),
    colors = breaks,
    shape = "circle"
  ) |>
  set_view(center = c(-77, 39), zoom = 8)

Rasters and a colorbar

A remote Cloud Optimized GeoTIFF is read directly by the browser, so the server must support CORS and HTTP range requests. add_colorbar() labels a single-band raster’s value range.

geolibre(layout = "maponly", height = 600) |>
  add_raster(
    "https://opendata.digitalglobe.com/events/california-fire-2020/pre-event/2018-02-16/pine-gulch-fire20/1030010076004E00.tif",
    name = "Imagery",
    bands = c(1, 2, 3)
  ) |>
  add_colorbar(colormap = "terrain", vmin = 0, vmax = 3000, label = "Elevation", units = "m")

Inspecting and rearranging layers

Every layer function addresses a layer by its name or its id. get_layers() returns one row per layer.

map <- geolibre() |>
  add_marker(-77.0369, 38.9072, name = "Capital") |>
  add_tile_layer("https://tile.openstreetmap.org/{z}/{x}/{y}.png", name = "OpenStreetMap")

get_layers(map)
#>                                     id          name    type visible opacity
#> 1 4a4de047-9a8b-4878-ac50-c030d9e36a92       Capital geojson    TRUE       1
#> 2 10fe6cbd-69ae-4c35-97a1-c79c2de4e2d3 OpenStreetMap     xyz    TRUE       1
#>                                           source features
#> 1                                           <NA>        1
#> 2 https://tile.openstreetmap.org/{z}/{x}/{y}.png       NA

map <- map |>
  move_layer("OpenStreetMap", 1) |>
  set_layer_opacity("OpenStreetMap", 0.5)

layer_names(map)
#> [1] "OpenStreetMap" "Capital"

describe_project() summarizes a project without printing its inlined data.

summary <- describe_project(map)
summary$layerCount
#> [1] 2
summary$mapView$zoom
#> [1] 2

Saving and exporting

A project saves to GeoLibre’s portable .geolibre.json format, which the web app, the desktop app, and the Python API all read. to_html() writes a standalone page that needs no running R session.

path <- file.path(tempdir(), "example.geolibre.json")
save_project(map, path)

restored <- geolibre(load_project(path))
layer_names(restored)
#> [1] "OpenStreetMap" "Capital"

html_path <- file.path(tempdir(), "example.html")
to_html(map, html_path, title = "Example map")
file.exists(html_path)
#> [1] TRUE

If the embedded application is unavailable, open GeoLibre Web directly.