#The packages need for shiny app
library(shiny)
library(lubridate)
library(tidyverse)
library(readr)
###The code needed for analysing the data frame of Ohio Analysis###
Ohio_data <- read_rds("data/statewide.rds")
Ohio_data <- filter(Ohio_data, !is.na(county_name), !is.na(subject_race))
Ohio_data <- Ohio_data %>%
  mutate(
    citation_issued = NA
  )
Ohio_data <-Ohio_data%>%
  mutate(year = year(date))

race_Ohio_data <- Ohio_data %>%
  group_by(subject_race) %>%
  summarise(
    arrest_rate = mean(arrest_made, na.rm = TRUE),
    search_rate = mean(search_conducted, na.rm = TRUE),
    warning_rate = mean(warning_issued, na.rm = TRUE)
  )

###The code needed for craeting the data frame for Spatical Analysis (Put this on the top,
###Because member say doesn't work put it after loading the package

Ohio_data$county_name <- gsub("\\s*\\w*$", "", Ohio_data$county_name)
Ohio_data$county_name <- toupper(Ohio_data$county_name)

race_rates <- Ohio_data %>%
    group_by(county_name, subject_race) %>%
    summarise(
        arrest_rate = mean(arrest_made, na.rm = TRUE),
        warning_rate = mean(warning_issued, na.rm = TRUE),
        search_rate = mean(search_conducted, na.rm = TRUE)
    )

#For generating the colopleth maps. helpers.R uses the maps and mapproj packages in R
library(maps)
library(mapproj)
# helpers.R is an R script that can help you make choropleth maps
library(plyr)
library(dplyr)
library(ggplot2)
library(rgeos)
library(rgdal)
library(maptools)
library(ggrepel)
library(scales)

      ###THe code need for Spatial Analysis(generating State maps)###
        oh_shp <- readOGR("data/REFER_COUNTY.shp")
      oh_shp <- fortify(oh_shp, region = "COUNTY") #shape file for map
      #The labels used in ggplot
      county_labels <- ddply(oh_shp, .(id), summarize, clat = mean(lat), clong = mean(long))
      #The data frame for producing balck and white maps
      race_rates_bw <- filter(race_rates, subject_race == "black" | subject_race == "white")
      race_rates_b <- filter(race_rates_bw, subject_race == "black")
      race_rates_w <- filter(race_rates_bw, subject_race == "white")
      rm(race_rates_bw)
      race_rates_bw <- merge(race_rates_b, race_rates_w, by = "county_name")
      race_rates_bw$arrest_rate_Delta <- race_rates_bw$arrest_rate.x - race_rates_bw$arrest_rate.y
      race_rates_bw$search_rate_Delta <- race_rates_bw$search_rate.x - race_rates_bw$search_rate.y
      race_rates_bw$warning_rate_Delta <- race_rates_bw$warning_rate.x - race_rates_bw$warning_rate.y



shinyApp(
    # Define UI for application that draws a histogram
  ui = tagList(
    shinythemes::themeSelector(),
    h2(textOutput("currentTime")),

  navbarPage(
    # theme = "cerulean",  # <--- To use a theme, uncomment this
    "Group3: Policing stop analysis",

    ##First module 01: "Columbus Analysis"
    tabPanel("Columbus analysis",

     sidebarPanel(
       selectInput("picture1", "Choose a pictire:",

         choices = c("Stop numbers each year in Columbus",
          "Stop counts/population in Columbus",
          "Outcome types and rates in Columbus",
          "White VS Black"),
         selected = "Stop numbers each year in Columbus" #default value
         ),

         # choices = c("1",
         #  "2",
         #  "3",
         #  "4")),

       h3("Modules was designed for the result of Wang Zhaohong's analysis."),
       h3("Racial discrimination in Columbus region."),
       actionButton("update", "Update View")

      ),#For sidebarPanel

     mainPanel(
       tabsetPanel(id = "module01",
         tabPanel("Stop numbers each year in Columbus",
          # value = "1",
           value = "Stop numbers each year in Columbus",
          img(src = "01.png", height = 720, width = 1280),
          hr(),
          h4("Discover: In this graph, we are compring the stop_count for each race,
            but since the larger proportion of white driver, so the comparison seems compatitive!")
          ),
         tabPanel("Stop counts/population in Columbus",
          # value = "2",
         value = "Stop counts/population in Columbus",
          img(src = "02.PNG", height = 720, width = 1280)
          ),
         tabPanel("Outcome types and rates in Columbus",
          value = "Outcome types and rates in Columbus",
          # value = "3",
          img(src = "03.PNG", height = 720, width = 1280)
          ),
         tabPanel("White VS Black",
          value =  "White VS Black",
          # value ="4",
          img(src = "04.PNG", height = 720, width = 1280)
          )#For last tabPanel
        )#For tabsetPanel
       )#For mainPanel
    ),#For tabPanel


    ##Second module 02: "Spatial Analysis"
    tabPanel("Difference in Black and White Analysis",

     sidebarPanel(
       selectInput("picture2", "Choose a pictire:",
         choices = c("arrest_rate", "search_rate", "warning_rate"),
         selected = "arrest_rate"#default value
         ),
       ##TD: Need explanation for spatial analysis(CJ)
       h3("This module is designed for result of Cornelius Johnson!"),
       h3("This result shows the spacial distribution of discrimination across the whole United States."),
       actionButton("update", "Update View")
      ),#For sidebarPanel

     mainPanel(
       tabsetPanel(id = "module02", ##The id must declared as Literal String
        tabPanel("arrest_rate",
            value = "arrest_rate",
            plotOutput(outputId = "plot21", height = "720", width = "1280")
          )

        )#For tabsetPanel
       )#For mainPanel
    ),#For tabPanel

    ##Thrid module 03: "Ohio Analysis"
    tabPanel("Ohio Analysis",

          sidebarPanel(
            "Outcome Type",
            selectInput("picture3", "Choose a type of rate: ",
              choices = c("arrest_rate", "search_rate", "warning_rate"),
              selected = "arrest_rate"
            )

          ##TD: Add description for drop down manu(CJ), don't forget to have a common beforehand

          ),#For sidebarPanel
        mainPanel(
          tabsetPanel(id = "module03",
            tabPanel("Type of rate",
                value = "arrest_rate",
                plotOutput(outputId = "plot31", height = "720", width = "1280")
              )

            # ,
            # tabPanel("search_rate",
            #   value = "search_rate",
            #    plotOutput(outputId = "plot32", height = "720", width = "1280")

            #   ),
            # tabPanel("warning_rate",
            #   value = "warning_rate",
            #    plotOutput(outputId = "plot33", height = "720", width = "1280")

            #   )

          )#For tabsetPanel

          )#For mainPanel
      ),#For tabPanel(Ohio analysis), end

    ##Fourth module 04: "Spatial Analysis Update"
    tabPanel("Spatial Analysis Update",

        sidebarPanel(
           selectInput("picture4", "Choose a type of race:",
             choices = c("asian", "black", "hispanic", "other" ,"white"),
             selected = "asian"#default value
             ),
            ##TD: add an explanation for your spacial analysis, and might what diferent race could do(CJ)
            actionButton("update", "Update View")
        ),#For sidebarPanel

        mainPanel(
          tabsetPanel(id = "module04",
            #Panel 1
            tabPanel("arrest_rate",
              value = "arrest_rate"
              ,plotOutput(outputId = "plot41", height = "720", width = "1280")
              ),
            # Panel 2
            tabPanel("search_rate",
              value = "search_rate"
              ,plotOutput(outputId = "plot42", height = "720", width = "1280")
              ),
            #Panel 3
           tabPanel("warning_rate",
              value = "warning_rate"
              ,plotOutput(outputId = "plot43", height = "720", width = "1280")
              )#End of all tabPanel
            )#For tabsetPanel
          )#For mainPanel

      )#For tabPanel(Spatial Analysis Update), end
    )#For navBarpage
  ),

  server = function(input, output, session) {





    ###First module 01: "Columbus Analaysis"
    observeEvent(input$picture1, {
        updateTabsetPanel(session, "module01",
                          selected = input$picture1
        )
    })

    ###Second module 02: "Difference Blakck and white Analysis"
    observeEvent(input$picture2, {
        updateTabsetPanel(session, "module02",
                          selected = input$picture2
        )
    })

    #for arrest_rate
    output$plot21 <- renderPlot({

      ##The switch button to change the value of input of y for ggplot
      rate <- switch(input$picture2,
          "arrest_rate" = race_rates_bw$arrest_rate_Delta,
          "search_rate" = race_rates_bw$search_rate_Delta,
          "warning_rate" = race_rates_bw$warning_rate_Delta
        )

        fill <- switch(input$picture2,
          "arrest_rate" = arrest_rate_Delta,
          "search_rate" = search_rate,
          "warning_rate" = warning_rate
          )

      })


    ##Third module 03: "Ohio State Analysis"
    #@input:
    ##plot1 is the output obejct from mainPanel;
    ##'input$picture3' is the selection made from user, arrest_rate, search_rate, warning_rate

    observeEvent(input$picture3, {
        updateTabsetPanel(session, "module03",
                          selected = input$picture3
        )
    })

    #for arrest_rate
    output$plot31 <- renderPlot({
      y <- switch(input$picture3,
          "arrest_rate" = race_Ohio_data$arrest_rate,
          "search_rate" = race_Ohio_data$search_rate,
          "warning_rate" = race_Ohio_data$warning_rate
        )
      title <- switch(input$picture3,
          "arrest_rate" = "Arrest Rate",
          "search_rate" = "Search Rate",
          "warning_rate" = "Warning Rate"
        )
    ggplot(race_Ohio_data, aes(x = subject_race, y, fill = subject_race)) +
    geom_bar(stat = "identity", position = position_dodge()) +
    labs(title = c("Racial Disparities in ", title ,"in Ohio"),
         x = "Driver Race",
         y = input$picture3,
         fill = "Driver Race")

      })

    ###"Forth Module: "Spatial Analysis update"###

    observeEvent(input$picture4, {
        updateTabsetPanel(session, "module04",
                          selected = input$picture4
        )

    })



    output$plot41 <- renderPlot({

        race <- switch(input$picture4,
          "asian" = "asian/pacific islander",
          "black" = "black",
          "hispanic" = "hispanic",
          "other" =  "other/unknown",
          "white" = "white"
          )
      # quantity <- filter(race_rates, subject_race == "hispanic")$arrest_rate
      # ggplot() + geom_map(data = filter(race_rates, subject_race == "hispanic"), aes(map_id = county_name, fill = arrest_rate),
      #                     map = oh_shp) +
      #     expand_limits(
      #         x = oh_shp$long -1,
      #         y = oh_shp$lat) +
      #     scale_fill_gradient2(
      #         low = "red",
      #         mid = "white",
      #         midpoint = median(race_rates$arrest_rate),
      #         high = scales::muted("blue"),
      #         limits = c(min(race_rates$arrest_rate), max(race_rates$arrest_rate)+.01)) +
      #     geom_text_repel( aes(x = clong, y = clat, label = id),
      #                      data = county_labels,
      #                      size = 2.25,
      #                      point.padding = NA) +
      #     theme_minimal() +
      #     theme(axis.text.x=element_blank(),
      #           axis.text.y = element_blank())+
      #     labs(title = "Selected Race Arrest Rate Distribution",
      #          fill = "Arrest Rate")

      ##Genearting  the graph

      # ggplot...
      #Code for testing ggplot
      # ggplot(data = mpg) +
      #     geom_point(mapping = aes(x = displ, y = hwy, color = class))
      })

    #Showing current time(Optional)
  output$currentTime <- renderText({
            invalidateLater(1000, session) #1000 milisecond = 1 sec
            paste("The current time is", Sys.time())
            })
  }#End of server
)#End of shinyApp