Shiny plot in ggplot not respecting "fill" argument in plot command

Question

I've included a lot of sample code because I'm new to Shiny (and still very beginner with R) so I'm not sure where the problem lies.

The first two code blocks are the ui.R and server.R of an app, along with a picture of the output. The problem I'm having is that the "fill" argument in the ggplot command isn't being respected, and I'm getting a mono-color output in the chart, when it should have two different colors in the histogram dependent upon the "id" value in the DF.

The last block of code and image show essentially the same code being run in R-Studio, with the desired effect achieved. You can see in the included output screenshot that there are two overlain histograms colored by the value of the "id" column.

Thanks for any help!

Example App:

server.R

library(shiny)
library(ggplot2)

id <- sample(0:1, 100, replace=T)
val <- sample(seq(1:25), 100, replace=T)
val2 <- sample(seq(1:10), 100, replace=T)

data <- data.frame(id, val, val2)

# Define a server for the Shiny app
shinyServer(function(input, output) {

  # Fill in the spot we created for a plot
  output$featurePlot <- renderPlot({

    # Render histogram

    # Note to readers, the outcome is the same whether the fill command
    # is as written, or written as: fill=as.factor(id)

    p <- ggplot(data, aes_string(x=input$feature, fill=as.factor(data$id))) +
        geom_histogram(alpha=0.5, aes(y=..density..), position='identity', binwidth = input$binSize);
    print(p)
  })
})

ui.R

library(shiny)
library(ggplot2)

id <- sample(0:1, 100, replace=T);
val <- sample(seq(1:25), 100, replace=T);
val2 <- sample(seq(1:10), 100, replace=T);

data <- data.frame(id, val, val2);

# Define the overall UI
shinyUI(

  fluidPage(

    titlePanel("why u no color?"),

    sidebarLayout(

      sidebarPanel(
        selectInput("feature", "Feature:",
                    choices=colnames(data),
                    selected='val'),
        hr(),
        helpText("Select a feature to compare"),

        sliderInput("binSize",
                    "Size of Bins",
                    min = 1,
                    max = 10,
                    value = 2)
    ),

    mainPanel(
      plotOutput("featurePlot")
    )
  )
)
)

app output:

Here is what I want

# Example Data For help

library(ggplot2)

id <- sample(0:1,100,replace=T)
val <- sample(seq(1:25),100,replace=T)
val2 <-sample(seq(1:10), 100, replace=T)

df <- data.frame(id, val, val2)

ggplot(df, aes(x=val, fill=as.factor(id))) +
  geom_histogram(alpha=0.7, aes(y=..density..), position = 'identity', binwidth=1)

You can run the code and see the desired output, but here's a picture just in case

Answer 1

You are setting your fill in a aes_string but you're not passing a string, you are trying to pass a vector. Your are mixing up aes() and aes_string(). Try separating them

p <- ggplot(data, aes_string(x=input$feature)) +
    geom_histogram(alpha=0.5, aes(fill=as.factor(id), y=..density..), 
        position='identity', binwidth = input$binSize);