R/visualise_dependencies.R

In baslat/sak: The Swiss Army Knife of R

#' Visualise function dependencies
#'
#' This function creates an interactive visualisation of function dependencies.
#' It can work on functions within a package, or functions loaded into your
#' global environment. Because the underlying functions are a bit weird, the
#' argument syntax isn't quite intuitive. The default behaviour will visualise
#' the function dependencies of what is in your global environment. If you want
#' to visualise the dependencies of a package, you need to pass the prefixed
#' name of a function from that package. Check the examples.
#'
#' @param .f (unquoted function name; default = \code{NULL}) an unquoted and
#'   prefixed function name
#'
#' @inheritDotParams foodwebr::foodweb
#'
#' @return a \code{visNetwork} object
#' @export
#'
#' @examples
#' \dontrun{
#'
#' # To visualise the dependencies of your global environment:
#' sak::load_custom_functions() # You need the functions in the env
#' visualise_dependencies()
#'
#' # To visualise the dependencies of a package
#' # Note if you are in a package repo first run `devtools::load_all()`
#' # You need to pass the name of a function from the package (any function will
#' # do)
#' visualise_dependencies(.f = sak::open_path)
#' }
visualise_dependencies <- function(.f = NULL, ...) {
 rlang::check_installed("foodwebr")
	rlang::check_installed("tidygraph")
	rlang::check_installed("visNetwork")

 fw <- foodwebr::foodweb(
		FUN = .f,
		filter = FALSE,
		...
	)

 tidy_fw <- tidygraph::as_tbl_graph(fw)

 # Styling the network
	main_col <- "lightblue"
	hover_col <- "darkblue"
	highlight_col <- "darkblue"

	# Displaying the network
	visNetwork::visIgraph(tidy_fw) %>%
		visNetwork::visOptions(
			highlightNearest = list(
				enabled = TRUE,
				hover = TRUE
			),
			nodesIdSelection = list(
				enabled = TRUE,
				main = "Select function"
			)
		) %>%
		visNetwork::visEdges(
			color = list(
				color = main_col,
				hover = hover_col,
				highlight = highlight_col
			),
			arrows = list(
				to = list(
					enabled = TRUE,
					scaleFactor = 1L
				)
			)
		) %>%
		visNetwork::visNodes(
			color = list(
				background = main_col,
				border = hover_col,
				hover = hover_col,
				highlight = highlight_col
			)
		)
}