Thread by @WeAreRLadies

👩🏼‍💻 How do you set up your own #ShinyApp?

It's easy in #rstats! Start a new #Rproject and select "Shiny Application". It will create a project with an "app.R" file for you ✨

Once it's open, you can replace the code that is already in the "app.R" file with this code snippet below👇 It does all the magic and shows how you can build a simple #ShinyApp 🔮

(it's here for you to try: bit.ly/shinyapp-test)

What the ShinyApp does:

You have checkboxes on the left side that let you choose countries (it's the ISO3 abbreviation, so "RWA" stands for Rwanda) and, depending on what you selected, your #ShinyApp will show a (non-realistic) population size for each country in a new plot.

When looking at the code, it might be overwhelming at first - but if we think of it as the pseudo-code snippets that I tweeted earlier, things start falling into place 🤓

But let's start from the beginning ⭐

We'll load the essential packages first - we obviously need {shiny} 📦 but I also load {dplyr} for data wrangling, {ggplot2} for visualization, and {overviewR} for the data (you can take any other data here).

In the next step, we load the toy data (toydata) from overviewR and now you're ready to go!

What you now see in code is how you can fill the UI 👤 and server 🧠 with content.

If you compare the pseudo code with the actual code, you'll see how similar they are. We define the title in "titlePanel()", add a "sidebarPanel()" (this is where the user selects what to see), and a "mainPanel()" (this is where the output lives).

For the sidebarPanel, I picked checkboxes for the selection in the sidebar of the UI 👤 (but there are multiple other possibilities for more control widgets): shiny.rstudio.com/tutorial/writt…).

To build the connection between the server 🧠 and the UI 👤 later, we'll do one important thing: we give our selector a unique ID: "countries" (you can see it in the code, but more on this once we're talking about the server 🧠)

For the mainPanel, we want a plot, so we call "plotOutput("first_plot")". "first_plot" in the parentheses refers to the plot that we later build in the server 🧠. Think of it as another ID or cross-reference that allows the server 🧠 and the UI 👤 to speak to each other.

For the server 🧠, I added a #ggplot2 inside the renderPlot() function to generate the visualization 👩🏼‍🎨

These render* calls are typical parts of the server 🧠. Just wrap them around your normal {ggplot2} calls (for instance).

And what you also see is that we store our visualization in "output$first_plot" 👈 this is where our "first_plot" is generated that we call later in the UI 👤. And there's something else - otherwise, your selection in the UI wouldn't work.

If you look at the filter, you'll see that we filter for countries in "input$countries" ("dplyr::filter(ccode %in% input$countries)"). This is exactly the "countries" ID that we added to the UI 👤 previously.

And now the magic happens 🧙🏻‍♀️

We call "shinyApp (ui = ui, server = server)" and bring the server 🧠 and the UI 👤 together.

And here's how it "runs" in action! Just highlight the code and let it run - a new window with your first ShinyApp will open 🙌

(I wanted to capture the full window so that you get the intuition of what it looks like in practice. That's why it's probably a bit small - but the code remains the same as above ✨ )

It's a really simple app and can be further customized - if you want to copy-paste the code to play around with it yourself, it's all in this repository: bit.ly/repo-shinyapp

Some things that I learned when building ShinyApps:

✨ Sketch your ShinyApp on paper (or tablet) with a pencil. This helps to understand what you want to get (and also what's probably missing). It's a bit like doing a design workshop with yourself.

✨ Start simple. Trust me, your ShinyApp will get complicated soon enough 😄

✨ 📦 {echarts4r} is a package for making beautiful visualizations (echarts4r.john-coene.com)

✨ If you're up for a production-ready ShinyApp, have a look at 📦 {golem}: engineering-shiny.org/golem.html (and the #Golemverse: golemverse.org)

✨ If you're not so much up for writing code, have a look at 📦 {shinyuieditor} - it allows you to create a UI without coding rstudio.github.io/shinyuieditor/…

✨ Deploying a ShinyApp on shinyapps.io is simple (if you don't have any data etc. restrictions) - just follow these steps: bit.ly/deploy-shiny

... and that all the talk about brains and bodies reminds me so much of Liv Moore 🧟‍♀️💜