How to use select input feature without having to specify the variable names of the data set

[gcefalu]

Hello,

I am trying to select input with the code below, and it works very well for a specific data set where I can I can use the data file labels in the selectInput feature. However, I do not know how to make the feature general so that I can ulpoad any data set and can get input without having to specify the names of all the variables. Also, the feature should allow for any data set with a different number of columns. I would appreciate help with this.

 selectInput("variable2", "Select variable:",
              c(
                 "Height" = "Height",
                 "Girth" = "Girth",
                 "Volume" = "Volume"
                #"cyl" = "cyl",
                #"mpg" = "mpg",
                #"disp" = "disp",
                #"hp"  = "hp",
                #"drat" = "drat",
                #"wt" = "wt",
                #"qsec" = "qsec",
                #"vs" = "vs",
               # "am" = "am",
               # "gear" = "gear",
               # "carb" = "carb"
              )),  

 output$data <- renderTable({
    
    #mtcars[, c(input$variable), drop = FALSE]
    trees[, c(input$variable), drop = FALSE]
  }, rownames = TRUE)
  

[prubin]

The shiny::updateSelectInput() function will let you update/replace the list of choices in a select input. You can put it inside an event observer so that it is triggered when the data set changes.

[gcefalu]

Thank you very much. Could you please provide an example of an event observer?

[prubin]

Try this small example.

library(shiny)

ui <- fluidPage(
  selectInput("df", "Select a data frame:", choices = data()$results[, "Item"], selected = "mtcars"),
  selectInput("var", "Select a variable:", choices = names(mtcars)),
  h4("Your choice:"),
  textOutput("choice")
)

server <- function(input, output) {
  # Listen for a variable selection and echo it to the text field.
  output$choice <- renderText({
    input$var
  })
  
  # Listen for a change of database and update the selection of variables.
  observeEvent(
    input$df,
    {
      updateSelectInput(inputId = "var", choices = get(input$df) |> names())        
    }
  )
}

# Run the application 
shinyApp(ui = ui, server = server)

^{Created on 2025-09-24 with reprex v2.1.1}

[margusl]

There's also varSelectInput() -- "select variables from a data frame",
it's used in a demo on https://shiny.posit.co/ landing page.

For your use case and with updateVarSelectInput(), perhaps something like this:

library(shiny)
library(bslib)

datasets <- list(mtcars = mtcars, trees = trees, penguins = penguins)

ui <- page_sidebar(
  sidebar = sidebar(
    radioButtons("inDs", "Select dataset:", choices = names(datasets)),
    varSelectInput("inVar", "Select variable:", datasets[[1]])
  ),
  tableOutput("outData")
)

server <- function(input, output, session) {
  ds <- reactive( datasets[[input$inDs]] )
  
  observe(updateVarSelectInput(session, "inVar", data = ds()))
  
  output$outData <- renderTable({ 
    var_char <- as.character(input$inVar)
    # deal with race conditions,  input$inVar might still include 
    # a column name from previously selected dataset
    req(var_char %in% names(ds()))
    ds()[1:12, var_char, drop = FALSE] 
  }, align = "l")
}

shinyApp(ui, server)

Screen recording

[gcefalu]

Thank you very much for your help both prubin and margusi.