How to convert Shiny app to a stand alone desktop application

Last updated on Aug 28, 2023 R

Last month I got a consultancy work where I had to examine data on male breast cancer in Indian men and create an application for my clients—who were essentially doctors—so that they could utilize it for their upcoming diagnoses. For performing fundamental statistical analysis, I typically use R programming, and I typically use Python to create apps. But this time, I faced difficulties because I needed to deploy the app quickly and all of the statistical tools that were utilized here weren’t available in Python (or, to be more honest, I didn’t know how to do that in Python).

The only choice left was to use Rshiny, but this time the issue was that they desired a desktop application. It was quite challenging for me at the time because I was only aware of how to utilize R Shiny to construct a web application that could be hosted on Shiny Server. I then began looking for answers, coming across a variety of them. In some articles, it is recommended to utilize Electron to transform a Siny app into a desktop application. Other sources advise utilizing the RInno package to do the same. But I found them all to be incredibly challenging. All of these approaches, which I attempted, were unsuccessful for me.

I later discovered “Packaging your Shiny App as a Windows desktop app” on Analytixware’s site, which is a wonderfully simple and efficient answer to my issue. Later, I came across Lee Pang’s article on R Bloggers where he offered a similar solution. Here I’m going to explain these steps to convert a Shiny app into a standalone desktop app. Note that, this steps are Windows specific, not for Mac.

Steps to convert Shiny App into a Desktop app:

Step 1

Create a Folder in a specific location add give it a name which you have decided for your App.

For example :

path: D:\Myapps\
Create New folder: D:\Myapps\MyApp1\

Step 2

Download:

R-Portable
Google Chrome Portable

and install both into MyApp1\ folder.

So inside MyApp1 folder there will be another two folder:

D:\Myapps\MyApp1\GoogleChromePortable\ D:\Myapps\MyApp1\R-Portable\

Step 3

Download all dependencies(R packages) of your shiny app into R portable.

Step 4

Create a folder called D:\Myapps\MyApp1\shiny\. This is where the files for your Shiny app (e.g. ui.R and server.R, data.csv, etc) will reside.

Step 5

Add the following to the server.R inside the shinyServer(function(input, output, session) { ... }). It is important to pass session as the third argument! The code you need to add is:

shinyServer(function(input, output, session)  { ... }) {
    # ... your other server code here
    
    # close the R session when Chrome closes
    session$onSessionEnded(function() { 
     stopApp()
     q("no") 
    })
    
    # ... your other server code here
  }

Step 6

To launch the application you will need two scripts:

D:\Myapps\MyApp1\runShinyApp.R : an R-script that loads the shiny package and launches your app via runApp() A shell script (either a .bat or .vbs file) that invokes R-portable.

Step 6.1 Create runshinyApp.R:

Open a new notepad file and paste this following lines of code and save it as runShinyApp.R on D:\Myapps\MyApp1\ location.

.libPaths("./R-Portable/App/R-Portable/library")
# the path to portable chrome
browser.path = file.path(getwd(),"GoogleChromePortable/GoogleChromePortable.exe")
options(browser = browser.path)
shiny::runApp("./Shiny/",port=8888,launch.browser=TRUE)

Step 6.2 Create shell script (run.vbs / run.bat):

Again open a new notepad file and paste this following code and save it as run.vbs or run.bat on D:\Myapps\MyApp1\ location.(I created run.vbs)

Randomize
CreateObject("Wscript.Shell").Run "R-Portable\App\R-Portable\bin\R.exe CMD BATCH --vanilla --slave runShinyApp.R" & " " & RND & " ", 0, False

Now, click on the run.vbs, you will see your App will be opened on the Google Crome browser.

I highly recommend you to read Analytixware’s blog and Lee Pang’s article on R Bloggers for clear understanding.

Blogs