############################ ui.R ##############################

library(shiny) library(ggplot2) library(nycflights13) library(data.table)

todat <- function(a,b,c) {paste(a,sprintf('%02d',b),sprintf('%02d',c),sep='/')}

shinyUI(fluidPage(

titlePanel("Old Faithful Geyser Data"),

sidebarLayout( sidebarPanel( sliderInput('dist','max distance', min = min(flights$distance), max = max(flights$distance), value = c(min(flights$distance),min(flights$distance)+300) ), selectInput('grp', 'group by', choices = c("origin","dest","carrier") ),

  selectInput('ddate1', 'dep date from', 
               choices = sort(unique(todat(flights$year,flights$month,flights$day)))
             ),
  selectInput('ddate2', 'dep date to', 
               choices = sort(unique(todat(flights$year,flights$month,flights$day)))
             )
  
),
# Show a plot of the generated distribution
mainPanel(
  tabsetPanel(
    tabPanel('Some statistics', dataTableOutput('tab1')),
    tabPanel('Distance stat', plotOutput("dist")),
    tabPanel('Air time stat', plotOutput("airt")),
    tabPanel('Dealy stat', plotOutput("dela")),
    tabPanel('Selected rows', dataTableOutput('tab2'))
  )
)

) ))

############################ server.R ##############################

library(shiny) library(data.table)

todat <- function(a,b,c) {paste(a,sprintf('%02d',b),sprintf('%02d',c),sep='/')} f <- flights f$dep_date <- todat(flights$year,flights$month,flights$day) setDT(f) f[, arr_time := paste(substr(sprintf('%04d',f$arr_time), 1, 2),substr(sprintf('%04d',f$arr_time), 3, 4),sep = ':')] f[, dep_time := paste(sprintf('%02d',f$hour),sprintf('%02d',f$minute),sep = ':')]

shinyServer(function(input, output) {

fl <- reactive({ f[dep_date >= input$ddate1 & dep_date <= input$ddate2 & distance >= input$dist[1] & distance <= input$dist[2], .(dep_date, dep_time, dep_delay, arr_time, arr_delay, carrier, tailnum, flight, origin, dest, air_time, distance )] })

output$dist <- renderPlot({ #ggplot(fl(), aes(fl()[,input$grp], distance)) + geom_boxplot() ggplot(fl(), aes(origin, distance)) + geom_boxplot() }) output$airt <- renderPlot({ ggplot(fl(), aes(origin, air_time)) + geom_boxplot() }) output$dela <- renderPlot({ ggplot(fl(), aes(origin, arr_delay)) + geom_boxplot() })

output$tab1 <- renderDataTable({ fl()[, .("flights no" = .N, "avg distance" = mean(distance, na.rm = TRUE), "avg air time" = mean(air_time, na.rm = TRUE), "avg delay" = mean(arr_delay, na.rm = TRUE), "max delay" = max(arr_delay, na.rm = TRUE) ), by = eval(input$grp)] })

output$tab2 <- renderDataTable({ fl()[,] })

})