Build your first dynamic shiny.semantic app using layout and grid

Create UI inside documents

This is how you generate your first UI with shiny.semantic:

uirender(
  shiny::tagList(
    h4("Numeric input", icon("world")),
    shiny.semantic::numeric_input("input", value = 0, label = ""),
    shiny.semantic::list_container(list(
        list(header = "Item 1", description = "My text for item 1", icon = "cat"),
        list(header = "Item 2", description = "My text for item 2", icon = "tree"),
        list(header = "Item 3", description = "My text for item 3", icon = "dog")
      ), is_divided = TRUE
    )
  ),
width = "200px", height = "280px")

Create a simple Shiny App

Let’s create a Shiny app with semanticPage() and just an h1 title.

ui <- semanticPage(
  title = "My first page",
  h1("My page")
)

server <- function(input, output, session) {}

shinyApp(ui, server)

Built-in layouts

Now let’s add sidebar layout sidebar_layout() with a sidebar panel sidebar_panel() and main panel main_panel(). The sidebar is displayed with a distinct background color and typically contains input controls. The main area occupies ⅔ of the horizontal width and typically contains outputs.

ui <- semanticPage(
  title = "My first page",
  h1("My page"),
  sidebar_layout(
    sidebar_panel(),
    main_panel()
    )
)

server <- function(input, output, session) {}

shinyApp(ui, server)

Next let’s fill in the main panel main_panel() with a Fomantic (Semantic previously) UI segment containing 2 cards.

ui <- semanticPage(
  title = "My first page",
  h1("My page"),
  sidebar_layout(
    sidebar_panel(),
    main_panel(
      segment(
         cards(
           class = "two",
           card(class = "red",
             div(class = "content",
                 div(class = "header", "Main title card 1"),
                 div(class = "meta", "Sub title card 1"),
                 div(class = "description", "More detail description card 1")
             )
           ),
           card(class = "blue",
             div(class = "content",
                 div(class = "header", "Main title card 2"),
                 div(class = "meta", "Sub title card 2"),
                 div(class = "description", "More detail description card 2")
                 )
             )
           )
         )
      )
    )
  )

server <- function(input, output, session) {}

shinyApp(ui, server)

Time to fill in sidebar sidebar_panel(), so let’s add a dropdown. We create it input using dropdown_input().

ui <- semanticPage(
  title = "My first page",
  h1("My page"),
  sidebar_layout(
    sidebar_panel(
      dropdown_input("mtcars_dropdown", c("mpg", "cyl", "disp", "hp"), value = "mpg"),
      textOutput("dropdown")
    ),
    main_panel(
      segment(
         cards(
           class = "two",
           card(class = "red",
             div(class = "content",
                 div(class = "header", "Main title card 1"),
                 div(class = "meta", "Sub title card 1"),
                 div(class = "description", "More detail description card 1")
             )
           ),
           card(class = "blue",
             div(class = "content",
                 div(class = "header", "Main title card 2"),
                 div(class = "meta", "Sub title card 2"),
                 div(class = "description", "More detail description card 2")
             )
           )
         )
      )
      )
    )
)

server <- function(input, output, session) {
  output$dropdown <- renderText(input$mtcars_dropdown)
}

shinyApp(ui, server)

Let’s make dropdown do something and generate a histogram and a simple plot from a selected mtcars column using dropdown.

ui <- semanticPage(
  title = "My first page",
  h1("My page"),
  sidebar_layout(
    sidebar_panel(
      p("Select variable for plots:"),
      dropdown_input("mtcars_dropdown", c("mpg", "cyl", "disp", "hp"), value = "mpg")
    ),
    main_panel(
      segment(
         cards(
           class = "two",
           card(class = "red",
             div(class = "content",
                 div(class = "header", "Main title card 1"),
                 div(class = "meta", "Sub title card 1"),
                 div(class = "description", "More detail description card 1")
             )
           ),
           card(class = "blue",
             div(class = "content",
                 div(class = "header", "Main title card 2"),
                 div(class = "meta", "Sub title card 2"),
                 div(class = "description", "More detail description card 2")
             )
           )
         )
      ),
      plotOutput("histogram"),
      plotOutput("plot")
      )
    )
)

server <- function(input, output, session) {
  output$dropdown <- renderText(input$mtcars_dropdown)
  output$histogram <- renderPlot(hist(mtcars[[input$mtcars_dropdown]]))
  output$plot <- renderPlot(plot(mtcars[[input$mtcars_dropdown]]))
}

shinyApp(ui, server)

There is more layouts that you can use. Just check the documentation of: splitLayout, flow_layout, vertical_layout and some more!

Grid templates

Finally let’s change the layout of the plots on the page and make them appear next to each other. We will use grid() for it and define a grid template using grid_template() with 1 row and 2 columns of the same size. Our grid template will contain 2 areas chart1 and chart2

grid_charts <- grid_template(
  default = list(areas = rbind(c("chart1", "chart2")),
                 rows_height = c("100%"),
                 cols_width = c("50%", "50%"))
)

We assign plots’ outputs plotOutput() to both areas inside the grid function.

This is the final effect of our work:

ui <- semanticPage(
  title = "My first page",
  h1("My page"),
  sidebar_layout(
    sidebar_panel(
      p("Select variable for plots:"),
      dropdown_input("mtcars_dropdown", c("mpg", "cyl", "disp", "hp"), value = "mpg")
    ),
    main_panel(
      segment(
         cards(
           class = "two",
           card(class = "red",
             div(class = "content",
                 div(class = "header", "Main title card 1"),
                 div(class = "meta", "Sub title card 1"),
                 div(class = "description", "More detail description card 1")
             )
           ),
           card(class = "blue",
             div(class = "content",
                 div(class = "header", "Main title card 2"),
                 div(class = "meta", "Sub title card 2"),
                 div(class = "description", "More detail description card 2")
             )
           )
         )
      ),
      grid(grid_charts,
           chart1 = plotOutput("histogram"),
           chart2 = plotOutput("plot")
         )
      )
    )
)

server <- function(input, output, session) {
  output$dropdown <- renderText(input$mtcars_dropdown)
  output$histogram <- renderPlot(hist(mtcars[[input$mtcars_dropdown]]))
  output$plot <- renderPlot(plot(mtcars[[input$mtcars_dropdown]]))
}

shinyApp(ui, server)

You can define grids for different screen sizes. Here is an example on how to define it for mobile.