arcgisviz

arcgisviz builds interactive charts with the ArcGIS Maps SDK for JavaScript. Charts render as htmlwidgets, so they work in the Viewer, in Quarto, and in Shiny.

Installation

pak::pak("R-ArcGIS/arcgisviz")

Usage

Start with arc_chart() and pipe. Each set_*() function maps one thing.

library(arcgisviz)

arc_chart(penguins) |>
  set_type("bar") |>
  set_x(species) |>
  set_y(body_mass) |>
  set_stat("mean") |>
  set_color(species) |>
  set_labs(title = "Palmer penguins", y = "Mean body mass (g)")

arc_bar(), arc_col(), arc_scatter(), and arc_line() are shorthand for the common pipelines.

arc_scatter(penguins, bill_len, bill_dep) |>
  set_color(body_mass, palette = "Blue 3")

A numeric column becomes a continuous gradient and a categorical one gets a colour per value. palette takes any of the 521 Esri colour ramps or a vector of R colours.

Axes

set_axis() handles one axis at a time. set_flipped() draws bars sideways.

arc_bar(penguins, island) |>
  set_axis("y", limits = c(0, NA), integer_only = TRUE) |>
  set_flipped()

Configuration

Every chart is backed by S7 classes mirroring the SDK’s WebChart JSON. Reach for them when you need something the set_*() functions do not cover.

library(arcgisviz)

chart <- arc_bar(penguins, species)

class(chart@webchart)
#> [1] "arcgisviz::WebChart" "S7_object"
chart@webchart@series[[1]]@type
#> [1] "barSeries"