ui.r

#
# Author:   Linnea Powell, Stephanie Wilcoxen, 
#           Ignacio Pezo, and Cristian Nuno
# Purpose:  Draft Dashboard
#

# Load necessary packages
library( ggthemes)
library( shiny )
library( shinydashboard )
library( leaflet )
library( geojsonio )
library( magrittr )
library( scales )
library( htmltools )
library( htmlwidgets )
library( DT )
library( dplyr )
library( stringr )
library( stringi )
library( MatchIt )
library( plotly )
library( censusapi )
library( ggplot2 )
library( igraph )
library( networkD3 )
library( rCharts )
library( pander )

# Import data from github function
source_github <- function( url ) {
  # load package
  require(RCurl)
  
  # read script lines from website and evaluate
  script <- getURL(url, ssl.verifypeer = FALSE)
  eval(parse(text = script), envir=.GlobalEnv)
} 
################# Importing External Objects ##############
# Import leaflet
leaflet_url <- "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Build%20App/leaflet_ny.r"
source_github( leaflet_url )

# Import table
fancy_table_url <- "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Build%20App/fancy_table.r"
source_github( fancy_table_url )

# import county overview SANKY
co_url <- "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Build%20App/aggregation_sanky_gra163.R"
source_github(co_url)

# load the census function using the RAW link
census_url <- "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Build%20App/census_comparator.R"
source_github( census_url )

# import small multiples aggregated data
agg_url <- "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Build%20App/small_multiples_aggregated_data.R"
source_github(agg_url)

############ Building the Dashboard##################

# A dashboard has 2 parts: a user-interface (ui) and a server

# The UI consists of a header, a sidebar, and a body.

# The server consists of functions that produce any objects
# that are called inside the UI

## customize header ##
header <- dashboardHeader(title = "USAspending Explorer"
                          , tags$li( a( href = "https://www.maxwell.syr.edu/deans/"
                                        , img( src = "https://github.com/USAspendingexplorer/USAspending-explorer/raw/master/Images/SUMaxell_Logo.png"
                                               , title = "Maxwell School Homepage", height = "30px")
                                        , style = "padding-top:10px; padding-bottom:10px;")
                                     , class = "dropdown"
                          ) # end of Maxwell Logo
                          , tags$li(a(href = "https://github.com/USAspendingexplorer/USAspending-explorer"
                                      , img( src = "https://github.com/USAspendingexplorer/USAspending-explorer/raw/master/Images/GitHub_Logo.png"
                                             , title = "GitHub", height = "30px")
                                      , style = "padding-top:10px; padding-bottom:10px;")
                                    , class = "dropdown"
                          ) # end of GitHub logo
) # end of header
## customize sidebar ##
sidebar <- dashboardSidebar(
  sidebarMenu(
    menuItem("Overview", tabName = "Over", icon = icon("home"))
    , menuItem("Explore", tabName = "Explore", icon = icon("search"))
    , menuItem("Compare", tabName = "Compare", icon = icon("bar-chart"))
    , menuItem("Use Case", tabName = "Use", icon = icon("briefcase"))
    #, menuItem("About", tabName = "About", icon = icon("users"))
  )
)

## customize body ##
body <- dashboardBody(
  # initialize tabs
  tabItems(
    
    
    ########################### 3. OVERVIEW TAB ###########################
    
    tabItem(tabName = "Over"
            , h2("Welcome to the USAspending Explorer")
            # , shiny::p("Under the U.S. Treasury Department’s leadership, the new site will allow taxpayers to examine", span( strong(" nearly $4 trillion in federal spending each year ")), "and see how this money flows from Congressional appropriations to local communities and businesses. The data is compiled by Treasury from federal agencies and published quarterly beginning in May 2017. The data will be updated each quarter.")
            # , shiny::p("Generally, federal spending can be separated between mandatory and discretionary spending. USAspending Explorer examines the discretionary side of federal spending by only look at spending transactions classified as grants. Given the open-source nature of our work, it is entirely possible to explore contracts, loans, and other financial assistance spending types.")
            # , shiny::p("All spending totals exclude grants which were directly award to the state government.")
            # , shiny::p("USAspending Explorer is an open source tool designed to assist in the navigation of federal data made publicly available through usaspending.gov. Through dynamic visualizations and search filters, the Explorer can answer the who, what, where, why and how of the complex federal government spending landscape.")
            , shiny::p("USAspending Explorer is an open source tool designed to assist in the navigation of federal data made publicly available through ", span( a("usaspending.gov", href = "https://www.usaspending.gov/DownloadCenter/Pages/DataDownload.aspx", target = "_blank")), ". Through dynamic visualizations and search filters, the Explorer can answer the who, what, where, why and how of the complex federal government spending landscape.")
            , shiny::p("This explorer serves as a tool to understand federal funding at the county level. It focuses on federal assistance in the form of grants and excludes grants that go directly to the state government because of the complexity of determining the final recipient of those funds.") 
            , shiny::p("With USAspending Explorer, you can understand how tax dollars are spent, explore who receives federal grants in your area, and see what that funding is directed towards. This tool shows total funding going to each county, and how funds are distributed geographically within the state adjusted by population. It also offers flexible comparisons between counties and the state average, displays county demographic information to inform county comparisons, and breaks down county funding by the funding agency, recipient, and program to better understand what is driving variations across counties.")
            , h3("All Federal Grant Funding Received by County, FY 2016")
            , fluidRow(
              column(width = 4
                     , DT::dataTableOutput("tbl")
              ) # end of column 1
              , column( width = 8
                        , leafletOutput("mymap", height=600)
              ) # end of column 2
            ) # end of row 1
    ) # end of Overview tab
    
    
    ########################### 2. EXPLORE TAB ###########################
    
    
    , tabItem( tabName = "Explore"
               , h2("County Explorer") # blank space
               , shiny::p("This diagram shows the flow of funding from funding agencies to recipients with widths proportional to the amount of funding. For simplicity it displays the top 10 funding agencies in each county and groups all other agencies into an other category. It includes only positive outlays.")
               , br()
               , fluidRow(
                 column( width = 2
                         , box( title = "County Exploration", status = "primary"
                                , solidHeader = TRUE, collapsible = FALSE, width = NULL
                                , selectizeInput(
                                  inputId='county',
                                  label='Select a county:',
                                  choices= c("NY State", sort(unique(gra16.3$county))),
                                  selected=c("NY State")
                                ) ) # end of box 1
                 ) # end of column 1
                 , column( width = 10
                           , box( title = "Flow of Funds from the Top 10 Agencies", status = "primary",
                                  solidHeader = TRUE, collapsible = FALSE, width = NULL
                                  , sankeyNetworkOutput( "sankey" ) ) # end of box 2
                 ) # end of column 2
               ) # end of row 2
               , fluidRow(
                 column( width = 12
                         , box( title = "Top Recipient Details", status = "primary"
                                , solidHeader = TRUE, collapsible = FALSE, width = NULL
                                , infoBoxOutput("top")
                                , infoBoxOutput("top.dollars")
                                , infoBoxOutput("top.num")
                         ) # end of box 3
                 ) # end of column 3
               ) # end of row 3
               , fluidRow(
                 column( width = 12
                         , box( title = "All FY16 Federal Grants by County", status = "primary"
                                , solidHeader = TRUE, collapsible = FALSE, width = NULL
                                , DT::dataTableOutput("countyTbl")
                         ) # end of box
                 ) # end of column
               ) # end of row
               
    ) # end of second tab
    
    ########################### 3. COMPARE TAB ###########################
    
    , # Third tab content
    tabItem(tabName = "Compare"
            , h2("County Comparison of Project Grant Funding")
            , shiny::p("In this section you can compare project grant funding patterns across counties. The following visuals build on the same federal grant data for New York state in FY 2016 but focus solely on project grants. Project grants are awarded for a specific purpose and based on the merit of the grant application, meaning that any variation from county to county is at least in part under the applicant’s control. This type of grant may be of more interest to users wishing to understand differences between counties than grants distributed based on a formula determined by law and often based on demographic information or less competitive block grants would be.")
            , shiny::p("The Compare Tab is divided into three sections: (1) Find Similar Counties; (2) Choose Counties to Compare; and (3) View Grant Funding Details.")
            
            
            ##################### 3.1 ROW ############################
            
            , h4( strong(""))
            , h4( strong("1. Find Similar Counties"))
            , shiny::p("Use the county demographics table and hover over the population vs poverty plot to find similar counties. Data is from ACS 2015.")
            , fluidRow(
              column( width = 7
                      , box( title = "Demographics for All Counties", status = "primary"
                             , solidHeader = TRUE, collapse = FALSE, width = NULL
                             , DT::dataTableOutput("censusTable")
                             ) # end of box 1
              ) # end of column 1
              , column( width = 5
                        , box( title = "Population vs. Poverty", status = "primary"
                               , solidHeader = TRUE, collapse = FALSE, width = NULL
                               , plotlyOutput("plotlyplot")
                        ) # end of box 2
              ) # end of column 2
            ) # end of row 1
            
            ##################### 3.2 ROW ############################
            
            , h4( strong("2. Choose Counties to Compare"))
            , shiny::p("Select the counties you want to compare and visualize their demographics and grant type structure.")
            , fluidRow(
              column( width = 2
                      , box( title = "Select Counties", status = "primary"
                             , solidHeader = TRUE, collapsible = FALSE, width = NULL
                             , selectizeInput(
                               inputId='your_county', 
                               label='Select up to 4 counties to compare:', 
                               choices= sort(unique(gra16.3$county)),
                               selected=c("Onondaga", "Herkimer", "Erie"), 
                               multiple = TRUE, 
                               options = list(maxItems = 4)
                             ) ) # end of box 1
              ) # end of column 1
              ,column( width = 5
                        , box( title = "Demographics for Selected Counties", status = "primary"
                               , solidHeader = TRUE, collapse = FALSE, width = NULL
                               , shiny::plotOutput("censusPlot", height = 500)
                        ) # end of box 2
              ) # end of column 2
              , column( width = 5
                        , box( title = "Grant Types for Selected Counties"
                               , status = "primary"
                               , solidHeader = TRUE, collapse = FALSE, width = NULL
                               , shiny::plotOutput("percapPlot", height = 500)
                        ) # end of box 3
              ) # end of column 3
            )
            
            ##################### 3.3 ROW  ############################
            
            , br()
            , h4( strong("3. View Grant Funding Details"))
            , shiny::p("Examine the project grant funding received by each county broken down by the recipient type and federal agency awarding the funds. This allows you to identify what areas or recipients are driving the variation in funds. You can further explore the specific grants and recipients that make up any of this variation in the data table below.")
            , fluidRow(
              column( width = 12
                      , box( title = "Federal Project Grant Funding by County, Agency, and Recipient (Per Capita)"
                             , status = "primary", solidHeader = TRUE, collapse = FALSE
                             , width = NULL
                             , shiny::plotOutput("smallMultiples", height = 1800)
                      ) # end of box 4
              ) # end of column 4
            ) # end of row 3
            
            ##################### 3.4 ROW ############################
            
            , fluidRow(
              column( width = 12
                      , box( title = "Federal Spending Details", status = "primary"
                             , solidHeader = TRUE, collapsible = FALSE, width = NULL
                             , selectInput(
                               inputId = 'recipient'
                               , label='Select a recipient:'
                               , choices = c( sort(unique(as.character(gra16.3$recip_cat_type))))
                               , selected = "Higher Ed"
                             ) # end of radio buttons
                             , selectInput(
                               inputId = 'maj'
                               , label='Select an Agency:'
                               , choices = c( sort(unique(as.character(gra16.3$maj_agency_cat))))
                               , selected = "Health And Human Services"
                             ) # end of radio buttons
                             , DT::dataTableOutput("cfdaTable")
                      ) # end of box 7
              ) # end of column 7
            ) # end of row 4
    ) # end of 3rd Tab Item
    
    
    ########################### 4. USE CASE TAB ###########################
    
    
    , tabItem( tabName = "Use"
               , h1("Welcome to the Use Case")
               , h2("User:")
               , shiny::p("Ms. Sherry Garcia. Public official at the Onondaga County’s Industrial Development Agency (OCIDA)")
               , h2("Interest:")
               , shiny::p("The county of Onondaga is alarmed by its rising fiscal deficit and the County Executive has asked her staff to look for new sources of revenue. Ms. Garcia, a public official at OCIDA, is interested in exploring publicly available federal spending data to better understand the structure of federal project grants coming into the county and to compare Onondaga to other local governments. Her main objective is to identify federal funding opportunities that the county could compete for by examining where counties similar to Onondaga have successfully found funds. Ms. Garcia is planning on using the USAspending Explorer tool to navigate through the data and produce a final report to the County Executive.")
               , h3( strong("1. Overview of Counties in New York State using the 'Overview' Tab") )
               , shiny::p("From the landing tab of USAspending Explorer, the user selects NY State and locates Onondaga on the map. The scaled colors in the county map allow the user to quickly compare Onondaga County’s total and per capita federal grant funding with the other counties in NY State. The data table allows the user a more detailed look.")
               , shiny::p("The current version of USAspending Explorer uses FY 2016 data and excludes all funds going directly to state government because these funds will likely be redistributed to other localities in the state. This allows the user to compare counties more meaningfully than if funds were reported as going to a single county when they’re actually redistributed to many.")
               , shiny::p("The user quickly finds that Onondaga is ranked 13th in total grants ($101,573,642) but 18th in per capita federal grants ($216.9) received. When looking at total grants, Onondaga is similar to Nassau ($107,817,884.25) and Cattaraugus ($97,653,885), however, Nassau receives much less on a per capita as compared to Onondaga while Cattaraugus receives much more. When looking only at per capita grants received, counties like St. Lawrence ($256.98) and Richmond ($296.13) are receiving similar amounts to Onondaga.")
               , shiny::p("These insights have given Ms. Garcia potential counties to compare Onondaga with.  To find out more about where funding is coming from and who is receiving it within Onondaga, she moves on to the next tab.")
               , h3( strong("2. Exploring the Flow of Funds using the 'Explore' Tab") )
               , shiny::p("The Flow of Funds diagram allows Ms. Garcia to see the relationship between funding agencies (to the left) and recipients (to the right) within Onondaga. She may also choose to see this diagram at the State level and for other counties.")
               , shiny::p("Ms Garcia quickly realizes that the Health and Human Services Agency (HHS) is the dominant funder of project grants for Onondaga County, and the money is primarily going into local higher education institutions and non-for-profits. Reflecting on this information, she realizes that the clear dominance of higher education institutions as recipients of federal grants in Onondaga county can be explained by the prominence of local higher education institutions like Syracuse University. Many questions arise for Ms. Garcia. Do other counties with big universities receive as much grant money as Onondaga? Are those grants coming from the same main funding agencies?")
               , shiny::p("Ms. Garcia also remembers that Rensselaer county was the top per capita grants recipient in NY State and decides to explore how grants are flowing in that particular county. She finds that HHS is again the main funding agency, however and contrary to Onondaga, HHS funds are delivered primarily to non-for-profit organizations in Rensselaer.")
               , shiny::p("St. Lawrence, which was close to Onondaga in per capita spending, has a very different structure. In St. Lawrence, Transportation is the main funding agency and government institutions are the main recipients of these funds. Why is St. Lawrence receiving more money from Transportation? How similar are St. Lawrence and Onondaga in terms of their demographics? What other similar counties should be explored?")
               , shiny::p("After establishing a general picture of funding in New York, the user hopes to compare what kind of federal funding exists within each county. They choose counties based on their own knowledge or county demographics. First, they examine how much of the federal funding going to each county is in the form of project grants (because these awards are assigned through an open competitive process). They use the variable assistance type to hone in on spending from project grants versus spending from other types of grants such as block or formula grants. They assume that counties who have a large amount of their funding coming from project grants are successfully finding and applying for competitive opportunities while counties with low project grant funding may be missing out on potential awards. The next step is to drill down the analysis for a better understanding of these awards.")
               , h3( strong( "3. Comparing Counties using the 'Compare' Tab" ) )
               , shiny::p("The County comparator tool offers Ms. Garcia key visuals that can help her to answer these questions. In this tab, the user finds a dynamic county demographic visualization, which helps her better understand how comparable the counties she is interested in examining are based on their population size, median household income and the percentage of the population which are living below the federal poverty level.")
               , shiny::p("Ms. Garcia is still able to refer to the amount of grants received by each county by looking at the per capita bar chart. This chart expands on what Ms. Garcia saw in the county map by also breaking funding down by project grants and other types of grants. As a County official, Ms Garcia knows that project grants are awarded through a competitive process and may be a source of new funding for Onondaga, while other types of grants are normally determined by formulas or other fixed methods.")
               , shiny::p("Due to their demographic similarities, Ms Garcia decides to explore the comparison between Onondaga and Orange county. Orange county is getting very similar amounts of project grants. To do a drill down comparison and see where any differences in funding between the two counties is coming from, the user looks at the multiple comparator graphs (located beneath the county demographic comparison and grants type barchart visuals.")
               , shiny::p("This visualization allows the user to look at how the funds flowing into each county are distributed between different awarding agencies and are given to different recipient types. Ms Garcia skims the multiple graphs and finds that the Government and Other recipients in Orange county is receiving a lot of money from the Department of Transportation. Onondaga is not receiving any significant funds from the Department of Transportation. In addition, the Department of Education is also providing Government institutions at Orange County with significant funds, while this agency is not funding Onondaga at all. Orange and Onondaga counties are receiving similar amounts of funds from the Health and Human Services Agency.")
               , shiny::p("Looking for more insights, Ms Garcia includes St. Lawrence county as another comparator. Ms. Garcia finds that St. Lawrence is also getting a lot of money from Transportation. Why are St. Lawrence and Orange county getting so much for Transportation?")
               , h3( strong( "4. Looking at Specific Project Grants" ) )
               , shiny::p("In order to understand why Orange and St. Lawrence counties are getting more grant funding from the Department of Transportation and why Orange county gets more funding from the Department of Education than Onondaga is, Ms Garcia needs to drill down further. To do so, Ms Garcia uses the Federal Spending Program table in this tab. This is a dynamic table that shows the specific grants going to each county and allows the user to filter by keywords in order to find specific funding programs.")
               , h4(em("Transportation Programs"))
               , shiny::p("After filtering through the grants, Ms Garcia finds out that the Department of Transportation is providing large sums of money to St. Lawrence and Orange counties for a Program described as “Airport improvement Program”, while Onondaga is not receiving nearly as much. While this could mean that the airports in Onondaga do not need as much improvement, it might also signify that Onondaga could more aggressively pursue this funding. See the screenshot below:"
                          , br()
                          , img( src = "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Images/UC_datatable_airpot.JPG"
                                 , height = 190, width = 700
                          ) # end of first screenshot
               ) 
               , h4(em("Education Programs"))
               , shiny::p("When looking at the grants provided by the Department of Education to Orange County, Ms Garcia finds that there has been a significant amount of funding provided to a specific School District in the county. However, she quickly realizes that this funding is tied to tax exempt land in the county and isn’t an opportunity for Onondaga to compete for funds. See screenshot below:"
                          , br()
                          , img( src = "https://raw.githubusercontent.com/USAspendingexplorer/USAspending-explorer/master/Images/UC_datatable_schools.jpg"
                                 , height = 400, width = 700
                          ) # end of second screenshot)
               ) # end of second long paragraph
               , h3( strong( "5. What was Learned" ) )
               , shiny::p("Through the navigation of USAspending Explorer, Ms Garcia has learnt more about the federal funding structure in NY State and Onondaga county. Comparing Onondaga with other counties has given her more context to assess the performance of Onondaga in competing for project grants and has also allowed her to identify specific project grants that are providing similar counties with significant amounts of funds.")
               , shiny::p("Naturally, this exploratory tool generates more questions than it answers, but Ms Garcia has learned a lot about federal funding in NY State, Onondaga and other counties. Better understanding the federal funding landscape will lead Ms Garcia to make further inquiries, research more about the specific project grants that she identified, and develop a plan for how to compete for those funds in the future. Ms Garcia could also contact the counties she has been looking at and ask them about program details now that she has a better picture of what funding she is interested in.")
               
    ) # end of 4th tab customization
  ) # end of Tab Items
) # end of dasboard body



#################################################################
## Shiny UI ##
ui <- dashboardPage(
  header
  , sidebar
  , body
)