Calculating correlation coefficient from user selected values

Question

As part of a larger Shiny app, I'm trying to filter (using the diamonds dataset as an example) by price to create a reactive data frame. Using this data frame, I would like to correlate price with a user-selected variable. The filter works as expected. I'm having trouble passing the selected variables from the filtered data frame to cor().

library(shiny)
library(tidyverse)

num_vars <- c("", "carat", "depth", "table", "price", "x", "y", "z")

ui <- fluidPage(
  sliderInput("price",
    "Price range",
    min = 300,
    max = 20000,
    value = c(10000, 12000)
  ),
  selectInput("var", "Variable to plot",
    choices = c("", "carat", "depth", "table", "x", "y", "z")
  ),
  textOutput("correlation")
)

server <- function(input, output, session) {
  df <- reactive(diamonds %>% filter(price > .env$input$price[1] &
    price < .env$input$price[2]))

  output$correlation <- renderText({
    df_local <- df()
    x <- df_local[[input$price]]
    y <- df_local[[input$var]]
    paste0(cor(x, y))
  })
}
shinyApp(ui = ui, server = server)

Answer 1

Try this

output$correlation <- renderText({
    req(input$var,input$price,df())
    df_local <- df()
    x <- df_local["price"] 
    y <- df_local[[input$var]]
    paste0(cor(x, y))
  })