6 Shiny basics

The purpose of this chapter is to guide you through various processes that you might need to go through to develop a Shiny R application.

6.1 Integrating Shiny and AWS data storage

This section aims to demonstrate how to utilise the Amazon Web Service (AWS) API to load data from an AWS bucket into the R environment. It then provides an example of how to use this protocol within a Shiny web application.

6.1.1 Install the `aws.s3` package

# install package
install.packages('aws.s3')

# load package
library(aws.s3)

6.1.2 Getting your unique AWS access keys - refer to Sonia for help with this step

From https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_access-keys.html:
- AWS access keys allow you to sign programmatic requests to the AWS CLI or AWS API.
- Access keys consist of two parts: an access key ID (for example, AKIAIOSFODNN7EXAMPLE) and a secret access key (for example, wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY).
- Like a user name and password, you must use both the access key ID and secret access key together to authenticate your requests.
- Manage your access keys as securely as you do your user name and password (i.e. include the .Renviron file in your .gitignore file if this is a project you will be storing in a GitHub repository).

6.1.3 Setting up your `.Renviron` file

Once you have your access key, secret access key and AWS default region (see step 1), this is the file that will store this information.
Assuming that you are in your working directory/ same directory as your Shiny app.R file, run the command touch ./.Renviron from the command line to generate the .Renviron file.
Then open this file, either on the command line, vi .Renviron, or interactively on Rstudio.
To edit on the command line, use hit ESC followed by i to insert text, then to save and close, ESC followed by :wq.
Add your access key, secret access key and AWS default region, for example:

Figure 6.1: .Renviron file example

Save and close the .Renviron file.

6.1.4 Loading AWS objects

The following command will load your access keys from your .Renviron file in an anonymised format, meaning you can put this in the body of your Shiny app.R file.

Sys.setenv("AWS_ACCESS_KEY_ID"=Sys.getenv("AWS_ACCESS_KEY_ID"),
           "AWS_SECRET_ACCESS_KEY"=Sys.getenv("AWS_SECRET_ACCESS_KEY"),
           "AWS_DEFAULT_REGION"=Sys.getenv("AWS_DEFAULT_REGION"))

Now you can use the AWS API to access remote files using the aws.s3 package.
The following example uses the s3read_using function and calls read.csv on the object specified by the object argument.
Other reader functions can be used, such as readr::read_csv or fst::read.fst.

# load aws.s3 library
library(aws.s3)

# read in csv from AWS s3 bucket 
obj1 <- aws.s3::s3read_using(read.csv, object = "s3://your-bucket/dataframe.csv")

6.1.5 Shiny/AWS example

Loading data from AWS in a Shiny app.R file

app.R is the core server/UI file you will have if you are building a Shiny app.
This example app.R file includes all the necessary code to load data from an AWS bucket

## app.R ##

# load libraries
library(shiny)
library(aws.s3)

# load access keys
Sys.setenv("AWS_ACCESS_KEY_ID"=Sys.getenv("AWS_ACCESS_KEY_ID"),
           "AWS_SECRET_ACCESS_KEY"=Sys.getenv("AWS_SECRET_ACCESS_KEY"),
           "AWS_DEFAULT_REGION"=Sys.getenv("AWS_DEFAULT_REGION"))

# load data from AWS bucket, assign to global 
obj1 <<- aws.s3::s3read_using(read.csv, object = "s3://your-bucket/dataframe.csv")

# define server function
server <- function(input, output) {
  # server code
}

# define UI function
ui <- fluidPage(
  # UI code
)

# run shinyApp function 
shinyApp(ui = ui, server = server)