Community

An R package bringing together a range of methods as part of a framework for collecting, documenting, analyzing, and presenting community data (relating to people in places or groups).

See the wiki for more about how the package and data sites work.

Installation

Download R from r-project.org, then install the package from an R console:

# install.packages('remotes')
remotes::install_github('uva-bi-sdad/community')

And load the package:

library(community)

Use Cases

Documenting a Dataset

# say you had some data like this
data <- data.frame(
  x = rnorm(100),
  y = rnorm(100),
  g = sample(letters[1:5], 100, TRUE)
)

# in a file like this
write.csv(data, "data.csv", row.names = FALSE)

# you would make a datapackage.json file to document it like this
init_data("example_data", path = "data.csv")

# or like this
init_data("example_data", overwrite = TRUE)
data_add("data.csv")

Making a Data Site

Data sites build off of data documentation to present datasets interactively.

This example creates a site to explore the mtcars dataset:

# initialize the site in a new directory
init_site("example_site", dir = "../example_site")

# use the created build.R file to prepare and add data
# the result should be a plain-text, tabular file with variables in columns
source("../example_site/build.R")

# finally, specify the site in the site.R file, then build it
site_build("../example_site", bundle_data = TRUE, open_after = TRUE)

See the Build a Data Site article for more.

JavaScript Testing

To run the JavaScript tests, first make sure Node.js is installed and a test site has been created from R:

init_site("test_site", dir = "../test_site")
source("../test_site/build.R")
site_build("../test_site")

Then install dependencies from a terminal in the package directory:

npm install

and run the tests, which will also rebuild the html report in the created coverage directory:

npm test