Box Plot in Shiny R

Question

I am trying to create/ learn an interactive box plot through Shiny, below id the code i m trying to use. This gives me the error

Warning: Error in model.frame.default: variable lengths differ (found for 'input$p')
[No stack trace available]

I am unable to figure this out, any help would be much appreciated

CODE:

library(shiny)

ui <- fluidPage(
  selectInput("p","p",choices = names(mtcars)),
  plotOutput("myplot"))

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

  output$myplot <- renderPlot({
    boxplot(mpg ~ input$p , data=mtcars)
  })
}

shinyApp(ui, server)

score 1 · Answer 1 · answered Oct 01 '18 at 15:35

boxplot expect boxplot(mpg ~ cyl , data=mtcars) while input$p will return a character vector as below

Browse[1]> input$p
[1] "mpg"

One solution is to use as.formula

library(shiny)

ui <- fluidPage(
  #use setdiff to avoid this Error 'Error in .subset2: attempt to select less than one element in integerOneIndex'
  selectInput("p","p",choices = setdiff(names(mtcars),"mpg")),
  plotOutput("myplot"))

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

  output$myplot <- renderPlot({
    m <- paste0('mpg','~',input$p)
    boxplot(as.formula(m) , data=mtcars)
  })
}

shinyApp(ui, server)

To get more explanation/insights see this question

Thanks A. Suliman, my grip on R is not really great, so for me your solution is kind of complicated. but i am trying to read and understand it — Fahadakbar, Oct 01 '18 at 16:54

score 1 · Accepted Answer · answered Oct 01 '18 at 15:41

1

Why dont you just use get

library(shiny)

ui <- fluidPage(
  selectInput("p","p",choices = names(mtcars)),
  plotOutput("myplot"))

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

  output$myplot <- renderPlot({
    boxplot(mpg ~ get(input$p) , data=mtcars)
  })
}

shinyApp(ui, server)

answered Oct 01 '18 at 15:41

Pork Chop

28,528
5
63
77

so that means , when ever i am to pick / refer to all the values in a named column in data frame, i need to use get function ? – Fahadakbar Oct 01 '18 at 16:57

score 0 · Answer 3 · answered Oct 16 '18 at 14:53

0

If you have a chance, maybe you want to check out library ggplot2. They have very very nice and easy to use functions and pretty plot.

answered Oct 16 '18 at 14:53

Y.Dong

36
8

Box Plot in Shiny R

3 Answers3