In saurfang/shinyCartogram: Cartogram map component for Shiny, using d3-cartogram

This R package wraps d3-cartogram in a reusable Shiny component. The concept and R Shiny binding implementation borrows heavily from leaflet-shiny.

cartogram.js is a JavaScript implementation of an algoritm to construct continuous area cartograms, by James A. Dougenik, Nicholas R. Chrisman and Duane R. Niemeyer, ©1985 by the Association of American Geographers.

This example replicates the original demo that Shawn Allen made for his d3-cartogram. This R package has made it possible for you to only focus on providing the dataset and column definitions rather than writing boilerplate codes that creates the map and associated visual updates.

Functions

Use the two functions cartogramOutput() and createCartogram() from your Shiny app to create cartogram maps.

# in ui.R
cartogramOutput(outputId, width = "100%", height = "500px",
                translateX = 0, translateY = 0, scale = 1,
                topojson = "cartogram/data/us-states-segmentized.topojson",
                colors = rev(brewer.pal(3, "RdYlBu")))

# in server.R
createCartogram(session, outputId)

TopoJSON

TopoJSON is an extension of GeoJSON that encodes topology. We use TopoJSON here to define boundaries of geographic regions and their names. You can find built-in .topojson files at system.file("www/data", package = "shinyCartogram"). There is also a Makefile that helps you convert .geojson files into .topojson.

Projections

Currently we use albersUsa as default map projection.

TODO: We intend to open up API to customize/update map projections. This allows you to arrange tiles in different ways. If you have such need, please let me know so we can better understand the best way to expose this ability.

Colors

Color string^[https://github.com/mbostock/d3/wiki/Colors] can be:

• rgb hexadecimal - "#ffeeaa"
• rgb shorthand hexadecimal - "#fea"
• named - "red", "white", "blue"

You can pass in:

• Single color to which maximum value will be associated. Minimum value will map to white.
• Two colors which correspond to minimum and maximum.
• Three colors where mid-point is either zero if possible or the arithmetic mean.

Methods

The following are methods that can be called on the map object that is created using createCartogram(). For example:

map <- createCartogram(session, "myMap")
map$setView(0, 0, 1.5)

Below is a list of currently available methods:

library(shinyCartogram)
res <- capture.output(str(unclass(createCartogram(list(output = list()), 'foo'))))
cat(gsub(':function ', '', res[-1]), sep = '\n')

Detailed documentation for all methods:

setView()

setView(x, y, scale, forceReset = FALSE)

Sets the transalte offsets and zoom level of the map. forceReset will completely reset the map view.

setData()

setData(data)

Sets the underlying dataset that cartogram reads scale and color variable from. data needs to be a data.frame and must contains a name column, which will be used to join with map tile ids. (e.g. state names)

setColumns()

setColumns(columns)

Sets the column definitions. columns needs to be a data.frame and must at least contain name column and can have optional title and format column which define the title and format used in tooltip.

columns are not mandatory and if not supplied title will defaults to column name and no number format will be applied.

Number Format

Number format must be character vector and can either contain

• A Javascript function that takes in a numeric value and output a string
• A format string for d3.format

scaleBy()

scaleBy(column)

Sets the column used to scale area.

If no scale column exists but a valid color column is selected, then original tile area size is preserved. In this case, this is equivalent to a choropleth map.

colorBy()

colorBy(column)

Sets the column used to scale color used to fill area.

If no color column exists but a valid scale column is selected, then that scale column will also be used to color map tile using its values.

Events

TODO: We plan to add click and hover events for map tiles

Data Sources

Population data is from US Census Bureau:

• http://www.census.gov/popest/data/state/totals/2013
• http://www.census.gov/popest/data/national/totals/2013/files/NST_EST2013_ALLDATA.csv

Location data is from Natural Earth:

• http://www.naturalearthdata.com/downloads/110m-cultural-vectors/
• http://www.naturalearthdata.com/http//www.naturalearthdata.com/download/110m/cultural/ne_110m_admin_1_states_provinces_lakes.zip
• http://vallandingham.me/shapefile_to_geojson.html

saurfang/shinyCartogram documentation built on May 29, 2019, 3:20 p.m.