Scatterplot in Shiny - r

Question

I am trying to create a scatterplot based on a csv I have loaded however when I run the code I either get no plot showing or an error when I include the aes mapping: "Mapping should be created with aes() or aes_()."
Can anyone give me pointers on where I am going wrong?
Code:
library(shiny)
library(ggplot2)
ui <- (fluidPage(
titlePanel("Pig Breeds"),
sidebarLayout(
sidebarPanel(
selectInput(inputId = "x",
label = "Pig Breeds:",
choices = c("total_pigs", "female_breeding_herd",
"in_pig_sows", "in_pig_gifts", "other_sows",
"maiden_gilts", "boars_for_service", "other_pigs"),
selected = "total_pigs"),
selectInput(inputId = "y",
label = "Year by year change:",
choices = c(2016, 2017, 2018, "year_on_year_change"),
selected = 2016),
actionButton(inputId = "update", label = "update")
),
mainPanel = (
plotOutput(outputId = "scatterplot")
)
)
)
)
server <- (function(input, output) {
output$scatterplot <- renderPlot({
ggplot(data=(read.csv("eu_pigs.csv")),
aes(x = output$x, y = output$y) +
geom_point())
observeEvent(input$update, {print(as.numeric(input$update))})
}
)
}
)
shinyApp(ui, server)

Answer 1

As the error message says, you’re using aes incorrectly. The function takes column names, not variable references. That is, replace
aes(x = output$x, y = output$y)
by
aes(x = x, y = y)
Or, more likely you want to be able to control the plot from the inputs, so you’d want to use
aes_string(x = input$x, y = input$y)
There are also quite a few stray parentheses and braces in your code. Remove those. Furthermore, mainPanel is a function that you need to call. Your code is instead assigning something to it.
And lastly, you actually need to plot your plot. After all these things are fixed, the relevant code looks like this:
ui <- fluidPage(
titlePanel("Pig Breeds"),
sidebarLayout(
sidebarPanel(…),
mainPanel(
plotOutput(outputId = "scatterplot")
)
)
)
server <- function(input, output) {
output$scatterplot <- renderPlot({
p = ggplot(data = read.csv("eu_pigs.csv")) +
aes_string(x = input$x, y = input$y) +
geom_point()
plot(p)
observeEvent(input$update, print(as.numeric(input$update)))
})
}
If the plot object is the last thing you are executing in the renderPlot function, you can omit plot:
output$scatterplot <- renderPlot({
ggplot(data = read.csv("eu_pigs.csv")) +
aes_string(x = input$x, y = input$y) +
geom_point()
})