9 Running STRAF on a local computer
9.1 Intro
STRAF is a web application, using the Shiny web application framework, based on the R programming language. It has been designed to work with a web browser and typically runs on a distant server.
There are several reasons for which one would prefer not using an application hosted on a web server: restrictions around the use of data, limited computational resources on the server, …
In this chapter, we describe a way to run STRAF on a local computer. It requires to install some dependencies (in particular, R and the straf R package) to be able to run STRAF locally.
9.2 Prerequisites
9.2.1 Installing R
You first need to install R, the language in which STRAF is programmed.
If your system is running under Windows, you can follow this link. You can download the installation software from there and then follow the instructions to install R on your system.
If your system is running under Mac OS, follow this link to install R
Finally, if you are using a Linux system,you should be able to install R using the using the following commands:
sudo apt update
sudo apt -y upgrade
sudo apt -y install r-base
Now, you should be able to run R-based software like STRAF on your computer.
9.2.2 Installing RStudio
RStudio is a free an open-source user-friendly environment built on top of R, that you can download from this website.
9.2.3 Install the shiny and straf packages
Now you can open RStudio and copy, paste and run the following commands in the R Console:
install.packages("remotes")
remotes::install_github("agouy/straf")
Now, the straf
R package should be installed on your system, which means you
are now able to run the app on your personal computer. You won’t need to re-run
the previous commands now that the package is present on your system.
9.3 Launching the application
You can now launch the application by copying and pasting the following commands in the console:
library(straf)
runStraf()
Et voilà! You should now have an open window. STRAF is running on your local computer, even though it uses a web browser, so the data you “upload” this way would in practice remain on your computer.