covid19dashboard.Rmd

---
title: "COVID-19: Cases Dashboard"
runtime: shiny
output: 
  flexdashboard::flex_dashboard:
    source_code: "https://github.com/smouksassi/covid19dashboard"
    social: [ "twitter", "facebook", "linkedin" ]
    orientation: rows
    vertical_layout: fill
    theme: spacelab
---
<style> .datatables{overflow: auto;}</style>

```{r global, include=FALSE}
library(rmarkdown)
library(tidyverse)
library(lubridate)
library(ggthemes)
library(scales)
library(knitr)
library(flexdashboard)
library(shiny)
library(shinyWidgets)
library(formattable)
library(plotly)
library(shiny)
library(DT)
library(sparkline)
library(htmlwidgets)
library(data.table)
library(jsonlite)
getDependency('sparkline')

staticRender_cb <-
  htmlwidgets::JS('function(){HTMLWidgets.staticRender();}')
repopath <-
  paste0(
    "https://raw.githubusercontent.com/CSSEGISandData/",
    "COVID-19/master/csse_covid_19_data/",
    "csse_covid_19_time_series/"
  )
jhu_url4 <- paste0(repopath, "time_series_covid19_confirmed_global.csv")
jhu_url5 <- paste0(repopath, "time_series_covid19_deaths_global.csv")
jhu_url6 <- paste0(repopath, "time_series_covid19_recovered_global.csv")
date  <- lubridate::today()
path4 <- file.path(".", paste0("confirmed_global", date, ".csv"))
path5 <- file.path(".", paste0("deaths_global", date, ".csv"))
path6 <- file.path(".", paste0("recovered_global", date, ".csv"))
repo = "https://api.github.com/repos/CSSEGISandData/COVID-19/"
path = "csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_global.csv&page=1&per_page=1"
lastcommitdata <- jsonlite::fromJSON(paste0(repo, "commits?path=", path))
lastcommitdate <- strptime(lastcommitdata$commit$author$date,"%Y-%m-%dT%H:%M:%S",tz = "UTC")

```

```{r, include=FALSE}
readdatadata <- eventReactive(input$loaddata,
                              {
                                withProgress(value = NULL, {
                                  setProgress(message = "Downloading data...")
                                  download.file(jhu_url4, path4)
                                  download.file(jhu_url5, path5)
                                  download.file(jhu_url6, path6)
                                  setProgress(message = "Parsing data...")
                                  list(
                                    read_csv(path4,  progress = FALSE),
                                    read_csv(path5,  progress = FALSE),
                                    read_csv(path6,  progress = FALSE)
                                  )
                                })
                              },
                              ignoreInit = FALSE,
                              ignoreNULL = FALSE)

dataprocessed <- reactive({
  req(readdatadata())
  withProgress(value = NULL, {
    setProgress(message = "computing stats...")
    confirmed <- readdatadata()[[1]] %>%
      rename(province = "Province/State",
             country_region = "Country/Region") %>%
      pivot_longer(
        -c(province, country_region, Lat, Long),
        names_to = "Date",
        values_to = "cumulative_cases"
      ) %>%
      mutate(Date = mdy(Date)) %>%
      arrange(country_region, province, Date) %>%
      mutate(province = ifelse(is.na(province), "all", province)) %>%
      group_by(country_region, province) %>%
      mutate(incident_cases = c(0, diff(cumulative_cases))) %>%
      ungroup()
    
    deaths <-  readdatadata()[[2]]  %>%
      rename(province = "Province/State",
             country_region = "Country/Region") %>%
      pivot_longer(
        -c(province, country_region, Lat, Long),
        names_to = "Date",
        values_to = "cumulative_deaths"
      ) %>%
      mutate(Date = mdy(Date)) %>%
      arrange(country_region, province, Date) %>%
      mutate(province = ifelse(is.na(province), "all", province)) %>%
      group_by(country_region, province) %>%
      mutate(incident_deaths = c(0, diff(cumulative_deaths))) %>%
      ungroup()
    
    recovered <-  readdatadata()[[3]]  %>%
      rename(province = "Province/State",
             country_region = "Country/Region") %>%
      pivot_longer(
        -c(province, country_region, Lat, Long),
        names_to = "Date",
        values_to = "cumulative_recovered"
      ) %>%
      mutate(Date = mdy(Date)) %>%
      arrange(country_region, province, Date) %>%
      mutate(province = ifelse(is.na(province), "all", province)) %>%
      group_by(country_region, province) %>%
      mutate(incident_recovered = c(0, diff(cumulative_recovered))) %>%
      ungroup()

    alldata <- left_join(confirmed, deaths)
    alldata <- left_join(alldata,recovered)
    alldata$countryprovince <-
      paste(alldata$country_region, alldata$province)
  })
  alldata <- alldata %>%
    filter(!str_detect(province, 'County|,'))%>%
    filter(!str_detect(province, 'recovered|Recovered'))
  alldata
})

colorlimits <- reactive({
  c(0, max(dataprocessed()$cumulative_deaths) + 10)
})
datasparkline <- eventReactive(input$computesparklines, {
  withProgress(value = NULL, {
    setProgress(message = "computing sparklines...")
    cases_sparkline_data <-  data.table(dataprocessed())
    keycols = c("countryprovince", "country_region", "province", "Date")
    setkeyv(cases_sparkline_data, keycols)
    cases_sparkline_data <- cases_sparkline_data[,
                                                 .(
                                                   cum_cases = spk_chr(cumulative_cases, type = "line"),
                                                   cum_death = spk_chr(cumulative_deaths, type = "line")
                                                 ),
                                                 by = .(countryprovince, country_region, province)]
    my_alldata <- merge(
      dataprocessed() %>%
        group_by(country_region, province, countryprovince) %>%
        slice(n()),
      cases_sparkline_data,
      by = c("countryprovince", "country_region", "province")
    )
  })
  my_alldata[, c("country_region", "province",
                 "cum_cases",
                 "cumulative_cases",
                 "cum_death",
                 "cumulative_deaths")]
}, ignoreInit = TRUE)
```

Filters {.sidebar }
=====================================
```{r}
column(12,
uiOutput("loadbutton"),
uiOutput("country"),
uiOutput("province"),
radioButtons("mapproj", "Map Type:",
             choiceNames = list(
               "Globe",
               "Map"
             ), 
               choiceValues = list(
                 "orthographic", "winkel tripel" 
               )),

div(
actionButton("computesparklines", "click for Sparklines/Mortality"),
style="vertical-align: bottom")
)
output$loadbutton <- renderUI({
actionBttn(inputId = "loaddata",
           label = HTML(paste0("Latest jhu Data:<br/>",
                               lastcommitdate," GMT<br/>",
                               "Click to Refresh")),
           color = "primary")
  })

output$country <- renderUI({
  shiny::req(dataprocessed())
  df <-  dataprocessed()
  if (is.null(df)) {
    return()
  }
  if (!is.null(df)) {
    selectInput(
      "Country",
      label = "Filter Countries",
      choices = levels(as.factor(df$country_region)),
      multiple = T
    )
  }
})
output$province <- renderUI({
  shiny::req(Data())
  shiny::req(input$Country)
  df <-  Data()
  if (is.null(input$Country))
    return()
  if (!is.null(input$Country)) {
    selectInput(
      "Province",
      label = "Filter Province/State",
      choices = levels(as.factor(df$province)),
      multiple = T,
      selected = levels(as.factor(df$province))
    )
  }
})
Data  <- reactive({
  req(dataprocessed())
  df <-   dataprocessed()
  if (is.null(input$Country)) {
    return(df)
  }
  if (!is.null(input$Country)) {
    df <- df %>% filter(country_region %in% input$Country)
  }
  df
})


Data2  <- reactive({
  req(Data())
  df <-   Data()
  if (is.null(input$Country)) {
    return(df)
  }
  if (is.null(input$Province)) {
    return(df)
  }
  if (!is.null(input$Country) && !is.null(input$Province)) {
    df <- df %>% filter(province %in% input$Province)
  }
  df
})

Data2last <- reactive({
  Data2() %>%
    group_by(country_region, province, countryprovince) %>%
    slice(n())
})
Data3  <- reactive({
  req(datasparkline())
  df <-   as.data.frame(datasparkline())
    if (is.null(input$Country)) {
    return(df)
  }
  if (!is.null(input$Country)) {
    df <- df %>% filter(country_region %in% input$Country)
  }
  if (is.null(input$Province)) {
    return(df)
  }
  if (!is.null(input$Country) && !is.null(input$Province)) {
    df <- df %>% filter(province %in% input$Province)
  }
  as.data.frame(df)
})
```


> Link to: <a href="https://github.com/CSSEGISandData/COVID-19" target="_blank">jhu data source repo </a>


Cases Map
====================================


Row {data-height=200}
-------------------------------------

### Total Cases {.value-box}

```{r}
output$vbox1 <- renderValueBox({
  df <-  Data2last()
  valueBox(sum(df[, "cumulative_cases"],na.rm=TRUE) ,
           icon = "fa-procedures", color = "primary")
})
valueBoxOutput("vbox1")
```

### Total Deaths {.value-box}

```{r}
output$vbox2 <- renderValueBox({
  df <-  Data2last()
  valueBox(sum(df[, "cumulative_deaths"],na.rm=TRUE) ,
           icon = "fa-calendar-times", color = "danger")
})
valueBoxOutput("vbox2")
```

### Total Recovered {.value-box}

```{r}
output$vbox3 <- renderValueBox({
  df <-  Data2last()
  valueBox(sum(df[, "cumulative_recovered"],na.rm=TRUE) ,
           icon = "fa-user-md", color = "success")
})
valueBoxOutput("vbox3")
```

Row {data-height=800}
-------------------------------------

### Map: Use Interactive Controls for Pan/Zoom/Hover {data-width=600} 

```{r}
output$outfig <- renderPlotly({ 
  req(Data2())
  g <- list(
  projection    = list(type = input$mapproj),
  fitbounds     = "locations",
  showland      = TRUE,
  showlakes     = FALSE,
  showrivers    = FALSE,
  landcolor     = toRGB("gray95"),
  subunitcolor  = toRGB("gray85"),
  countrycolor  = toRGB("black"),
  countrywidth  = 0.5,
  subunitwidth  = 0.5,
  showsubunits  = TRUE,
  showcountries = TRUE,
  resolution = 110,
  lonaxis    = list(
    showgrid = TRUE,
    gridwidth = 0.5,
    dtick = 5
  ),
  lataxis    = list(
    showgrid = TRUE,
    gridwidth = 0.5,
    dtick = 5
  )
)
  plot_geo(Data2last(), lat = ~ Lat, lon = ~ Long) %>% add_markers(
    text = ~ paste(
      country_region,
      ifelse(province == "NA", "", province),
      paste0("Cases:", cumulative_cases),
      paste0("Deaths:", cumulative_deaths),
      sep = "<br />"
    ),
    symbol = I("square"),
    size = ~ cumulative_cases ,
    color = ~ (cumulative_deaths + 0.0001),
    hoverinfo = "text"
  ) %>%
    colorbar(title = "Cumulative Deaths<br /> ",
             limits = colorlimits()) %>%
    layout(
      title = paste('Cumulative Covid-19 Cases on:', today()),
      geo = g
    )
})
plotlyOutput("outfig") 
```

### Raw Cases Numbers Over Time For Selected Countries/Regions: {data-width=600} 

```{r}
output$lineplot = renderPlotly({
  req(Data2())
  df <-   Data2() %>% ungroup()
  df$country_region <- factor(df$country_region)
  df$country_region <- reorder(df$country_region, - df$cumulative_deaths)
  df$country_region <- factor(df$country_region,
                              labels =
                                stringr::str_wrap(levels(df$country_region), 12))

  df$`Country Province`<-df$countryprovince
  df$`Cumulative Cases`<- df$cumulative_cases
  p <- ggplot(df,
              aes(x = Date,
                  y = `Cumulative Cases`,
                  group=`Country Province`)) +
    geom_line() +
    theme_bw() +
    labs(y = "Cases", x = "Date")
  ggplotly(p) 
})
plotlyOutput("lineplot")
```

Row {data-height=400}
-------------------------------------------- 

### Sparklines: Selected Rows are shown on the Mortality Plot {data-width=600}

```{r}
uiOutput("dynamicsparktable")
output$dynamicsparktable <- renderUI({ 
  if (input$computesparklines) {
output$sparktable<- DT::renderDataTable({
  req(Data3())
  df<- as.data.frame(Data3())
  print(names(df))
  spk_add_deps(
    datatable(
      df %>% 
        arrange(desc(cumulative_cases),
                desc(cumulative_deaths)),
      filter = 'none',
      escape = FALSE,
      selection = list(
        target = "row",
        mode = 'multiple',
        selected = c(1,2,3,4,5)
      ),
      extensions = c('Buttons','Select'), #c('Select', 'Buttons')
      options = list(
        dom = 'lfrtiBp',
        pageLength = 5 ,
        lengthMenu = list(c(5, 10, 15,-1), c('5', '10', '15', 'All')),
        #select = list(style = 'os', items = 'row'),
        #buttons = c('selectAll', 'selectNone', 'selectRows'),
        autoWidth = FALSE,
        scrollX = TRUE,
        scrollY = TRUE,
        target = 'row',
        drawCallback   = staticRender_cb,
        fnDrawCallback = staticRender_cb,
        #order = list(list(4, 'desc'), list(6, 'desc')),
        columns = list(
          list(title = ''),
          list(title = 'Country'),
          list(title = 'Province/State'),
          list(title = 'Cases Over Time'),
          list(title = 'Total Cases'),
          list(title = 'Deaths Over Time'),
          list(title = 'Total Deaths')
        )
      )
    )
  )
})
dataTableOutput("sparktable")
}
  else NULL
})
```

### Mortality Percentages: Total Deaths/Total Cases {data-width=600} 

```{r}
output$mortalityratioplot = renderPlotly({
  req(Data3()) 
    df <-  as.data.frame(Data3()) %>% 
        arrange(desc(cumulative_cases),
                desc(cumulative_deaths))%>%
    mutate(MortalityRatio = ifelse(
      cumulative_cases > 0,
      100 * (cumulative_deaths / (cumulative_cases)),
      0
    ))
  df$country_region <- factor(df$country_region)
  df$country_region <- reorder(df$country_region, - df$MortalityRatio)
  df$country_region <- factor(df$country_region,
                              labels =
                                stringr::str_wrap(levels(df$country_region), 12))
  s1 = input$sparktable_rows_selected  # rows on the current page
  df <- df[if( nrow(df)< max(s1) ) c(1:nrow(df)%in%s1) else s1 , , drop = FALSE]
  df$Country <- df$country_region
  df$Province<-df$province
  df$`Mortality Ratio`<- df$MortalityRatio
  
  p <- ggplot(df,
              aes(x = Country,
                  group = Province,
                  y = `Mortality Ratio`)) +
    geom_col(position = position_dodge()) +
    theme_bw() +
    labs(y = "Mortality Ratio", x = "Country/Region")
  
  ggplotly(p) #layout(xaxis = list(range = c(1,5)))
  
  
})
plotlyOutput("mortalityratioplot")
```

Data
====================================

Datatables {.tabset}
------------------------------------


### All Data 

```{r}
DT::renderDT ({
  print(head(as.data.frame(Data2()[, ])))
  
  datatable(
    as.data.frame(Data2()[, 1:11]),
    filter = 'top',
    extensions = 'Buttons',
    options = list(
              dom = 'lBfrtip',
        pageLength = 10 ,
        lengthMenu = list(c(5, 10, 15,-1), c('5', '10', '15', 'All')),
        buttons = list(
          list(
            extend = "csv",
            text = "Download Current Page",
            filename = paste0("pagedata",today()),
            exportOptions = list(modifier = list(page = "current"))
          ),
          list(
            extend = "csv",
            text = "Download Full Results",
            filename = paste0("fulldata",today()),
            exportOptions = list(modifier = list(page = "all"))
          )
        ),
      autoWidth = FALSE,
      scrollX = TRUE,
      scrollY = TRUE,
      columns = list(
        list(title = ''),
        list(title = 'Province/State'),
        list(title = 'Country'),
        list(title = 'Lat'),
        list(title = 'Long'),
        list(title = 'Date'),
        list(title = 'Cumulative Cases'),
        list(title = 'Incident Cases'),
        list(title = 'Cumulative Deaths'),
        list(title = 'Incident Deaths'),
        list(title = 'Cumulative Recovered'),
        list(title = 'Incident Recovered')
      )
  )
  )
},server = FALSE)
```