categorical data visualization problem in shiny

Question

I want to visualize the data categories, but I can't get what I want with the code below. When I want to divide the data into categories, it evaluates it as a whole. I can't get the chart categorized. What am I doing wrong?

library(shiny)
library(ggplot2)

ui <- fluidPage(

titlePanel("shiny demo"),

 sidebarLayout(

   sidebarPanel(
  
  fileInput(inputId = "file",
            label = "choose file",
            multiple = F),
  
  selectInput("p","variable1",
              choices = c("sex","smoke","happines","city","work")),
  
  selectInput("q", "variable2",
              choices = c("hopefulnes","happines")))
,
mainPanel(
  tabsetPanel(
    
    tabPanel(title = "Data",tableOutput("data_out")),
    
    tabPanel(title = "Bar",plotOutput("barp"))
    
      )
    )
  )
)
server <- function(input,output,session) {

 data <- reactive({
 req(input$file)

 df=read.csv(input$file$datapath,sep = ";",header = T)

 })
 output$data_out <- renderTable({
 data() 
 })
  
 output$barp <- renderPlot({      
 ggplot(data(), aes(input$p, ..count..)) + geom_bar(aes(fill = input$q), position = "dodge")  
 })

}
shinyApp(ui,server)

Data;

id  sex   smoke happines     hopefulness   work city
1   man   yes   very happy   very hopeful   yes   A
2   man   no    very happy   very hopeful   yes   A
3   man   no    unstable     not hopeful    no    C
4   woman no    unstable     not hopeful    yes   A
5   woman no    unstable     not hopeful    yes   B
6   man   yes   very happy   hopeful        yes   C
7   woman yes   happy        unstable       no    D
8   man   yes   not happy    not hopeful    yes   A
9   woman no    not happy    unstable       yes   B
10  man   no    very happy   very hopeful   yes   D

error code;

Input to asJSON(keep_vec_names=TRUE) is a named vector. In a future version of jsonlite, this option will not be supported, and named vectors will be translated into arrays instead of objects. If you want JSON object output, please use a named list instead. See ?toJSON.

thank for help

Answer 1

selectInput() is setup to output a string, whereas aes() is setup to expect a named column. If you select "smoke" from input$p and "happiness" from input$q , this gets fed into the function as "smoke" and "happiness" , and not smoke and happiness . Consequently, it's going to plot as if you sent this function:

ggplot(df, aes("smoke", ..count..)) +
  geom_bar(aes(fill = "happiness"), position = "dodge")

To handle the string output from selectInput() , you should use aes_string() in place of aes() . You'll need to change ..count.. to "..count.." , but it will work as if you asked to evaluate the following code chunk:

ggplot(df, aes_string("smoke", "..count..")) +
  geom_bar(aes_string(fill = "happiness"), position = "dodge")

The other way you can do this is to maintain using aes() , but just evaluate the string from input$p and input$q as a variable name. You can use get() to do that normally, so I believe it will also work in your app. You don't have to change ..count.. to "..count.." in this case:

# yields the same plot as above
ggplot(df, aes(get("smoke"), ..count..)) +
  geom_bar(aes(fill = get("happiness")), position = "dodge")