#
# This is a Shiny web application. You can run the application by clicking
# the 'Run App' button above.
#
# Find out more about building applications with Shiny here:
#
#    http://shiny.rstudio.com/
#

library(shiny)
library(ggplot2)

# Define UI for random distribution app ----
ui <- fluidPage(
  
  sidebarPanel(
      # Input: Select the random distribution type ----
      radioButtons("dist", "Distribution type:",
                   c("Normal" = "norm",
                     "Unif" = "unif",
                     "Binomial" = "binom",
                     "Poisson" = "pois",
                     "Chi-square" = "chisq",
                     "F" = "f",
                     "T" = "t")
                   ),
      sliderInput("n",
                  "Sample Size",
                  min = 1,
                  max = 10000,
                  value = 50),
      sliderInput("bins",
                  "Histogram bin count:",
                  min = 1,
                  max = 300,
                  value = 30),
      sliderInput("bw",
                  "Density Bandwidth",
                  min = 0, 
                  max = 1,
                  value = 0.5),
  
  # Normal distribution parameters
  conditionalPanel(
    condition = "input.dist == 'norm'",
    sliderInput("mean", 
                "mean",
                min = -100,
                max = 100,
                value = 0),
    sliderInput("sdev",
                "sdev",
                min = 0,
                max = 100,
                value = 1)
  ),
  
  # Normal distribution parameters
  conditionalPanel(
    condition = "input.dist == 'unif'",
    sliderInput("range", 
                "Minimum and Maximum",
                min = -100,
                max = 100,
                value = c(0, 10)
    )
  ),
  

  
  conditionalPanel(
    condition = "input.dist == 'binom'",
    sliderInput("size", 
                "Number of trials:",
                min = 1,
                max = 10000,
                value = 1),
    sliderInput("prob",
                "Probability of success:",
                min = 0,
                max = 1,
                value = 0.5)
  ),
  
  conditionalPanel(
    condition = "input.dist == 'pois'",
    sliderInput("lambda", 
                "Lambda (Mean occurrence per interval):",
                min = 0,
                max = 1000,
                value = 10)
  ),
  
  conditionalPanel(
    condition = "input.dist == 'chisq'",
    sliderInput("df", 
                "Degrees of freedom:",
                min = 1,
                max = 100,
                value = 2)
  ),
  
  conditionalPanel(
    condition = "input.dist == 'f'",
    sliderInput("df1", 
                "Degrees of freedom 1:",
                min = 1,
                max = 100,
                value = 1),
    sliderInput("df2",
                "Degrees of freedom 2:",
                min = 1,
                max = 100,
                value = 1)
  ),
  
  conditionalPanel(
    condition = "input.dist == 't'",
    sliderInput("df", 
                "Degrees of freedom:",
                min = 1,
                max = 100,
                value = 2)
  )),
  
  mainPanel(
    
    # Output: Tabset w/ plot, summary, and table ----
    plotOutput("histPlot")
    )
)

# Define server logic required to draw a histogram
server <- function(input, output) {
  output$histPlot <- renderPlot({
    
    random_sample <- data.frame(switch(input$dist,
                                       'norm' = rnorm(n = input$n, mean = input$mean, sd = input$sdev),
                                       'unif' = runif(n = input$n, min = input$range[1], max = input$range[2]),
                                       'binom' = rbinom(n = input$n, size = input$size, prob = input$prob),
                                       'pois' = rpois(n = input$n, lambda = input$lambda),
                                       'chisq' = rchisq(n = input$n, df = input$df),
                                       'f' = rf(n = input$n, df1 = input$df1, df2 = input$df2),
                                       't' = rt(n = input$n, df = input$df)
    ))
    
    quant_vect <- seq(min(random_sample), max(random_sample), (max(random_sample) - min(random_sample))/ 100)
    plot_pdf <- data.frame(switch(input$dist,
                                       'norm' = dnorm(quant_vect, mean = input$mean, sd = input$sdev),
                                       'unif' = dunif(quant_vect, min = input$range[1], max = input$range[2]),
                                       'binom' = dbinom(round(quant_vect), size = input$size, prob = input$prob),
                                       'pois' = dpois(round(quant_vect), lambda = input$lambda),
                                       'chisq' = dchisq(quant_vect, df = input$df),
                                       'f' = df(quant_vect, df1 = input$df1, df2 = input$df2),
                                       't' = dt(quant_vect, df = input$df)
    ))
    
    pdf_frame <- data.frame(quant_vect, plot_pdf)
    names(pdf_frame) <- c('quants', 'probs')
    
    names(random_sample) <- c('x')

    ggplot(data = random_sample, aes(x = x)) +
      geom_histogram(aes(y=..density..), bins = input$bins) +
      geom_vline(xintercept = mean(random_sample$x), color = 'red', linetype = 'dotted', size = 1) +
      geom_density(bw = input$bw) +
      geom_line(aes(x = quants, y = probs), color = 'red', data = pdf_frame)
      
  })
}

# Run the application 
shinyApp(ui = ui, server = server)