Validate that variables has expected number of levels

Usage

validate_n_levels(x, min_levels = 1, max_levels = 12, var_name)

Arguments

x

variable name. If x is not a factor, the unique values are treated as levels.

min_levels

cutoff for minimum number of levels of x

max_levels

cutoff for maximum number of levels of x

var_name

name of variable being validated for use in validation message

Details

If the number of levels of x is less than min_levels or greater than max_levels the validation will fail. This function is a wrapper for shiny::validate.

Examples

data <- data.frame(
  one = rep("a", length.out = 20),
  two = rep(c("a", "b"), length.out = 20),
  three = rep(c("a", "b", "c"), length.out = 20),
  four = rep(c("a", "b", "c", "d"), length.out = 20),
  stringsAsFactors = TRUE
)
ui <- fluidPage(
  selectInput(
    "var",
    "Select variable",
    choices = c("one", "two", "three", "four"),
    selected = "one"
  ),
  verbatimTextOutput("summary")
)

server <- function(input, output) {
  output$summary <- renderText({
    validate_n_levels(data[[input$var]], min_levels = 2, max_levels = 15, var_name = input$var)
    paste0(
      "Levels of selected treatment variable: ",
      paste(levels(data[[input$var]]),
        collapse = ", "
      )
    )
  })
}
if (interactive()) {
  shinyApp(ui, server)
}