Shiny Stage 4 - Dynamic Dashboards

Last updated on 2 min read

Dashboards extend the functionality of shiny. Shiny uses Bootstrap for layout whereas Shinydashboard uses AdminLTE, which is a theme built on top of Bootstrap. Here is a basic template - every dashboard must have the header, sidebar and body. However, these can be hidden if necessary.

library(shiny)
library(shinydashboard)

ui <- dashboardPage(
  dashboardHeader(),
  dashboardSidebar(),
  dashboardBody()
)

server <- function(input, output){

}
shinyApp(ui, server)

Before we go any further, I should note, the docs are here.

OK, so Shinydashboard has analogous layout controls to shiny, for example fluidRow and it uses box as the main container for elements. Let’s add in some of the functionality from the Shiny 101 post.

library(shiny)
library(shinydashboard)
ui <- dashboardPage(
  dashboardHeader(),
  dashboardSidebar(),
  dashboardBody(
    fluidRow(
      box(width = 8, plotOutput("plot1", height = 250)),
      box(width = 4,
          title = "Control",
          sliderInput("slider", "Observations", 
                      min = 1, max = 100, value = 50)
        
      )
    )
  )
)

server <- function(input, output){
  output$plot1 <- renderPlot({
    n <- input$slider
    hist(rnorm(n))
  })
}
shinyApp(ui, server)

In addition to box there is infoBox and valueBox for presenting particular forms of data to the user.

We can place a menu in the sidebar using the sidebarMenu function in the UI. Within sidebar we define menuItems each of which corresponds to a tabItem in the dash board body. Note that the icons are obtained from Font Awesome, which ships with Shiny.

library(shiny)
library(shinydashboard)
ui <- dashboardPage(
  dashboardHeader(),
  dashboardSidebar(
    sidebarMenu(
      menuItem("First", tabName = "first", icon = icon("dashboard")),
      menuItem("Second", tabName = "second", icon = icon("th"))
    )
  ),
  dashboardBody(
    tabItems(
      tabItem(tabName = "first", tags$h2("First content")),
      tabItem(tabName = "second", tags$h2("Second content"))
    )
  )
)

server <- function(input, output){
  output$plot1 <- renderPlot({
    n <- input$slider
    hist(rnorm(n))
  })
}
shinyApp(ui, server)

We will talk about Headers later.

Leaflet

This enables you to work with maps a little more easily. Leafletjs is a javascript library for working with interactive maps and you can interface that library using the CRAN leaflet package. You use it like this:

library(leaflet)
leaflet() %>%
  addTiles() %>%
  setView(-93.65, 42.0285, zoom = 17) %>%
  addPopups(-93.65, 42.0285, 'Stats Dept')

You can use the leaflet output within a Shiny app by calling leafletOutput("map") from the UI code and then using renderLeaflet() passing it the code that is shown above.

27 mins

https://www.rstudio.com/resources/webinars/dynamic-dashboards-with-shiny/

shiny R visualisation

Related