The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

pkgGraphR

Use Case

When developing large packages or Shiny apps in R, it can be difficult to track where a modification to one function might propagate down-stream. As a simple example, suppose you’ve moved from the initial prototype phase into a more serious development phase for a package and realize that you would like to change parameter names in a function from some.param to someParam in order to match other function parameterization. This function might live in the utils.R file and be called by functions spread out across several other .R files. Having a map (graph) of what functions depend on this function can be very helpful in making sure that all usages of this function get adjusted correctly.

As this package is specifically intended for development phase, it doesn’t require the package to be built and in fact works with any directory containing R files, or even a single ’.R; file. This has the advantage of being useful outside of R package development, specifically in cases such as shiny apps which are often developed outside of the package context.

Usage

There are 2 (or 3) steps to using pkgGraphR, first (optional) collect the function assignments with collectFunNames, next build the graph object (a list containing nodes and edges) with buildPackageGraph, finally visualize the results as desired with plotPackageGraph. The example below shows how to use each function assuming you are in the package or app directory you want to visualize.


library(pkgGraphR)

funclist <- collectFunNames(x = ".")
funcgraph <- buildPackageGraph(x = ".", 
                               unique.edges = TRUE, 
                               only.connected = FALSE)
# under default parameters, only the graph is required
plotPackageGraph(graph = funcgraph)
# alternatively, plot with grouping and/or coloring (requires fun.list)
plotPackageGraph(graph = funcgraph, 
                 fun.list = funclist, 
                 use.subgraphs = T, 
                 use.colors = T)

Known “issues”

There are a few known issues which should be taken into consideration.

  1. grViz doesn’t allow . in node names so if you use my.function be aware that grViz will show these as myfunction.

  2. Very large packages (e.g. dplyr) will be difficult to visualize. As a workaround, you can use htmlwidgets and webshot to generate a high resolution pdf as below.


p <- plotPackageGraph(graph = funcgraph)
htmlwidgets::saveWidget(p, "test.html")
webshot::webshot(url = "test.html", file = "test.pdf")

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.