Site icon R-bloggers

Fetch API Results from the Browser and send them to Shiny

[This article was first published on Colin Fay, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

How to fetch API results from client-side in a Shiny app and send them to R.

Why on earth?

Good question! Why on earth would you want your Shiny application to make API requests from the User Interface (i.e. from the browser)? Right now, if your application makes API calls, chances are that you have being doing them straight from R, and that it works pretty well. But in some cases it might not be the correct implementation, notably if the API limits requests based on an IP and your application gets a lot of traffic. For example, Brewdog’s PUNK API limits to one call per IP per second. In that scenario, if the calls are made from the server, you will only be able to serve one call per second, and if your users do a lot of API call at the same time, they will be slowed down, as the server’s IP is limited to 1 call per second. On the other end, if the API call is made from the users’ browsers, the server IP is no longer limited: each user has its own limitation.

How to

You can write API calls using the fetch() JavaScript function. It can then be used inside a Shiny JavaScript handler, or as a response to a DOM event (for example, with tags$button("Get Me One!", onclick = "get_rand_beer()"), as we will see below).

Here is the general skeleton, that would work when the API does not need authentication and returns JSON.

< !-- end list -->
// JAVASCRIPT FUNCTION DEFINITION
const get_rand_beer = () => {
    // FETCHING THE API DATA
    fetch("https://api.punkapi.com/v2/beers/random")
      // DEFINE WHAT HAPPENS WHEN JAVASCRIPT RECEIVES THE DATA
      .then((data) =>{
        // CONVERT THE DATA TO JSON
        data.json().then((res) => {
          // SEND THE JSON TO R
          Shiny.setInputValue("beer", res, {priority: 'event'})
        })
        // DEFINE WHAT HAPPENS WHEN THERE IS AN ERROR TURNING DATA TO JSON
        .catch((error) => {
          alert("Error catchin result from API")
        })
      })
      // DEFINE WHAT HAPPENS WHEN THERE IS AN ERROR FETCHING THE API
      .catch((error) => {
        alert("Error catchin result from API")
      })
  };
< !-- end list -->
observeEvent( input$beer , {
  # Do things with input$beer
})

And here it is, when the users click on the button, the API calls will be made from their browser, and then sent back to the server.

Note that the data shared between R and JavaScript is serialized to JSON, so you will have to manipulate that format once you receive it in R.

See an example at ColinFay/punkapi

To leave a comment for the author, please follow the link and comment on their blog: Colin Fay.

R-bloggers.com offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.