app.R

library(shiny)
library(shinyWidgets)
library(tidyverse)
library(sf)
library(leaflet)
library(DT)
library(leafpop)
library(FSA)
library(knitr)
library(ggtext)
# library(periscope)

# Module code
plotDownloadUI <- function(id, height = 400) {
  ns <- NS(id)
  tagList(
    fluidRow(
      column(
        2, offset = 10,
        downloadButton(ns("download_plot"), "Download plot")
      )
    ),
    fluidRow(
      plotOutput(ns('plot'), height = height)
    )
    
  )
}

plotDownload <- function(input, output, session, plotFun) {
  output$plot <- renderPlot({
    plotFun()
  })
  
  output$download_plot <- downloadHandler(
    filename = function() {
      "plot.png"
    },
    content = function(file) {
      ggsave(file, plotFun(), width = 9, height = 6)
    }
  )
}

# Read in summary fish metrics
metrics <- read_csv('standardized_fish_data.csv') %>%
  relocate(area) %>%
  relocate(N, .after = last_col()) %>%
  mutate(gcat = case_when(gcat == "Stock-Quality" ~ "S-Q",
                          gcat == "Quality-Preferred" ~ "Q-P",
                          gcat == "Preferred-Memorable" ~ "P-M",
                          gcat == "Memorable-Trophy" ~ "M-T",
                          gcat == "Trophy" ~ "T") %>%
           factor(levels = c("S-Q", "Q-P", "P-M", "M-T", "T"))) %>% 
  mutate(method = str_replace_all(method, " ", "_"), 
         waterbody_type = str_replace_all(waterbody_type, " ", "_"), 
         metric = str_replace_all(metric, "CPUE distance", "CPUE")) %>% 
  filter(method %in% c("boat_electrofishing", "raft_electrofishing", "gill_net_fall", "gill_net_spring", "drifting_trammel_net", "large_catfish_hoopnet", "bag_seine", "stream_seine", "backpack_electrofishing", "tow_barge_electrofishing"))

# Develop vectors of unique entries
uni.type <- c("North America", "Ecoregion", "State/Province")
uni.area <- sort(unique(metrics$area))
uni.spp <- sort(unique(metrics$common_name))
uni.method <- sort(unique(metrics$method))
uni.watertype <- sort(unique(metrics$waterbody_type))
uni.metric <- c("Length Frequency", "Relative Weight", "CPUE") 

# Read in ecoregions shapefiles
ecoregions <- read_sf("ecoregions1/NA_CEC_Eco_Level1.shp")
ecoregions_crs <- "+proj=laea +lat_0=45 +lon_0=-100 +x_0=0 +y_0=0 +a=6370997 +b=6370997 +units=m +no_defs"
ecoregions_trans <- ecoregions %>% 
  rmapshaper::ms_simplify() %>% 
  st_set_crs(ecoregions_crs) %>%
  st_transform("+proj=longlat +datum=WGS84")

# Read in lat/longs
# Lat/longs are the locations of sampling sites for datasets included in the summary metrics.
# They are used for the plot on the "Explore" page of the app and otherwise do not affect functionality.
# They are not included in this repo for data privacy reasons.
# To get the app to run, read in the `toy_locs.csv` file to populate this data object with fake data.

# # Read in lat/longs
locs <- 
  read_csv("sites.csv") %>%
  # read_csv("toy_locs.csv") %>%
  select(-date) # not parseable as is

# Read in example user upload data
ex <- read_csv("example_user_upload_data.csv")

# Create a "not in" function
`%nin%` <- negate(`%in%`)

#Shiny App
ui <- navbarPage(title = "AFS Standard Sampling App", 
                 theme = bslib::bs_theme(bootswatch = "sandstone"),
                 tabPanel("About",
                          wellPanel(
                            tags$head(tags$style(
                              type="text/css",
                              "#pics img {display: block; margin-left: auto; margin-right: auto; max-height: 90%; height: 90%; max-width: 100%, width: auto}"
                            )),
                            imageOutput("pics"),
                            h2("About Standard Sampling", align = "center"),
                            h4("Standardization of sampling methods allows fisheries professionals to compare data across large spatial and temporal scales, encourages data sharing and improves communication. 
In light of these benefits, the American Fisheries Society published the first edition of Standard Methods for Sampling North American Freshwater Fishes in 2009. 
A new edition of these methods was published September 2024.  
The goals of this project were to:"),
                            br(),
                            h4("(1) recommend standardized freshwater fish sampling methods for North America, and"),
                            h4("(2) provide an online database of fish data collected using AFS recommended standard methods where data from individual water bodies could be compared to existing rangewide, ecoregion, and state standards."),
                            br(),
                            h4("Since publication, numerous fisheries professionals have adopted the standard methods and many have noted the database as an important tool in management. 
The database allows for comparison of fish metrics commonly used in management to assess population health including growth, condition, length-frequency, and catch per unit effort data collected using standard methods.
When developing version two of the database, we examined the strengths and weaknesses of methods used for requesting, analyzing, and displaying data in an online format.
We used this information to achieve our goals of maximizing use and providing simple means to update this program for comparison of fisheries data."),
                            br(),
                            h4("We hope that these methods can be adopted by others, particularly those in data-poor regions, to maximize the ability to compare data. "),
                            h2("Details About App Usage", align = "center"), 
                            h4("Here, you can compare your data from an individual water body,", strong("collected using a standard AFS method"), "to averages across the range of the species, ecoregion, and statewide. This tool will help you assess if fish you have collected are high, average or low for the metric in question."),
                            h4("We collected thousands of data sets on fishes from across Canada, the United States, and some from Mexico to build comparison summaries. A ", strong("data set (", .noWS = "after"), em("N"), strong(")", .noWS = "before"), " is defined as data collected with AFS standard gears and methods during routine monitoring programs of an entire fish community or a population of specific fish ", strong("in a single waterbody conducted once during a year"), ". For example, data collected from Bass Lake (a small standing water body) sampled by boat electrofishing for Largemouth Bass in 2014 would represent 1 data set. In some cases, data sets included multi-day surveys of the same waterbody (e.g., large reservoirs) where the effort was summed across all sampling days. This excludes surveys targeting specific size groups or those with other biases (e.g., egg counts, juveniles fish surveys)."), 
                            h4("You will see the following tabs in the bar at the top of the page: "), 
                            br(), 
                            h4("- ", strong("About", .noWS = "outside"), ": gives an overview about the standard sampling webtool"), 
                            h4("- ", strong("Explore", .noWS = "outside"), ": displays desired subsets of the standard data in a map and a table"), 
                            h4("- ", strong("View", .noWS = "outside"), ": displays desired subsets of the standard data in graphical form"), 
                            h4("- ", strong("Compare Your Data", .noWS = "outside"), ": allows you to compare your data from a waterbody to standard data in rangewide, ecoregion, or state summaries"), 
                            br(), 
                            h4("Most of the app components can be translated into another language by going to ", 
                               a("Google Translate", href="https://translate.google.com/?sl=auto&tl=en&op=websites", .noWS = "outside"), 
                               " and entering the URL for the app, then selecting the desired language. Maps and plot can only currently be generated using the original version of the app. "), 
                            br(), 
                            h4(strong("If you would like the text of the app bigger, you can increase the font size by hitting the control and plus (+) keys on Windows, or command and plus (+) keys on Macs.")), 
                            br(), 
                            h2("Moving Forward", align = "center"), 
                            h4("These data sets were collected and compiled by ", 
                               a("Scott Bonar's lab", href="https://azcfwru.wixsite.com/azcfwru", .noWS = "outside"), 
                               ". Only summaries can be shared. Individual data sets cannot be shared with others because of legal restrictions on their use. As funding permits, additional data will be added to this app, including for more collection methods."), 
                            h4("All feedback on this app is greatly appreciated, and can be provided by sending an email to ", 
                               a("Dr. Bonar", href="mailto:SBonar@ag.arizona.edu", .noWS = "outside"), 
                               " or by submitting an ", 
                               a("issue", href="https://github.com/cct-datascience/AFS_database_code/issues", .noWS = "outside"), 
                               " through the project's ", 
                               a("GitHub repository", href="https://github.com/cct-datascience/AFS_database_code", .noWS = "outside"), 
                               ", which is also where the code is located."), 
                            h4("Developed in collaboration with the University of Arizona ",
                               a("CCT Data Science", href="https://datascience.cct.arizona.edu/"), "team."),
                            h4 ("Sponsored by:"), 
                            imageOutput("logo")
                          )
                 ),
                 
                 tabPanel(title = "Explore",
                          sidebarLayout(
                            sidebarPanel(
                              "Welcome to the American Fisheries Society Standard Sampling Database App!", 
                              br(),
                              br(),
                              "To explore standard fish data, select one or more options from each menu below.",
                              radioGroupButtons(inputId = "typechoice",
                                                label = "Show data by:",
                                                choices = uni.type,
                                                direction = "vertical",
                                                selected = "North America"),
                              uiOutput("dyn_area"),
                              selectInput(inputId = "sppchoice",
                                          label = "Select species:",
                                          choices = c("All", uni.spp),
                                          multiple = TRUE,
                                          selected = 'Bluegill'),
                              selectInput(inputId = "methodchoice",
                                          label = "Select method(s):",
                                          choices = c("All", uni.method),
                                          multiple = TRUE,
                                          selected = "boat_electrofishing"),
                              selectInput(inputId = "watertypechoice",
                                          label = "Select waterbody type(s):",
                                          choices = c("All", uni.watertype),
                                          multiple = TRUE,
                                          selected = "large_standing_waters"),
                              selectInput(inputId = "metricchoice",
                                          label = "Select metric(s):",
                                          choices = uni.metric,
                                          multiple = TRUE,
                                          selected = uni.metric),
                              downloadButton("filterdownload", "Download filtered"),
                              downloadButton("alldownload", "Download all")
                            ),
                            
                            mainPanel(
                              tabsetPanel(
                                tabPanel("Map", 
                                         br(), 
                                         h6("Points show approximate locations of data, and colored areas indicate", 
                                            strong("EPA Ecoregions Level I"), 
                                            ". Display name of ecoregion at a location by clicking on map. See more details on the ", 
                                            a("EPA website", href="https://www.epa.gov/eco-research/ecoregions-north-america", .noWS = "outside"), 
                                            "."), 
                                         br(), 
                                         leafletOutput("plotSites",
                                                       width = 700,
                                                       height = 500),
                                         textOutput("report_absent1"),
                                         textOutput("report_absent2")), 
                                tabPanel("Table", 
                                         DTOutput("filtertable"))
                              )
                            )
                          )
                 ), 
                 
                 tabPanel(title = "View",
                          sidebarLayout(
                            sidebarPanel(
                              "Welcome to the American Fisheries Society Standard Sampling Database App!", 
                              br(), 
                              br(),
                              "To view plots of standard fish data, select one option from each menu below.",
                              radioGroupButtons(inputId = "typechoice2",
                                                label = "Show data by:",
                                                choices = uni.type,
                                                direction = "vertical",
                                                selected = "North America"),
                              uiOutput("dyn_area2"),
                              selectInput(inputId = "sppchoice2",
                                          label = "Select species:",
                                          choices = uni.spp,
                                          selected ='Bluegill'),
                              selectInput(inputId = "methodchoice2",
                                          label = "Select method:",
                                          choices = uni.method,
                                          selected = "boat_electrofishing"),
                              selectInput(inputId = "watertypechoice2",
                                          label = "Select waterbody type:",
                                          choices = uni.watertype,
                                          selected = "large_standing_waters")
                            ),
                            mainPanel(plotDownloadUI("LF_plot"),
                                      p("Proportional size distribution length frequency. Black lines indicate standard error."),
                                      hr(),
                                      plotDownloadUI("RW_plot"),
                                      p("Relative weight by proportional size distribution categories. Points indicate means and lines indicate standard error."),
                                      hr(), 
                                      plotDownloadUI("CPUE_plot", height = "200px"), 
                                      p("Catch per unit effort. The box represents the middle 50% of the standard data with the median value indicated by the line inside. The whiskers extend to the smallest and largest values within 1.5 times the inter quartile range and any individual points outside are outliers. CPUE units depend on collection method: boat and raft electrofishing are fish per hour, gill net is fish per net nights, drifting trammel net is fish per 100-m drift, large catfish hoopnet is fish per 24 hour set, bag seine is fish per 0.25 arc (small standing waters) or 0.5 arc (rivers), stream seine is fish per 10-15m haul, and backpack and tow barge electrofishing are fish per 100m².")
                            )
                          )
                 ),
                 tabPanel(title = "Compare Your Data",
                          sidebarLayout(
                            sidebarPanel(
                              "Welcome to the American Fisheries Society Standard Sampling Database App!",
                              br(),
                              br(),
                              "Upload your fish data for comparison here: ", 
                              fileInput("upload", NULL,
                                        buttonLabel =  "upload",
                                        multiple = FALSE, 
                                        accept = (".csv"), 
                                        placeholder = ""), 
                              "To view plots of standard fish data, select one option from each menu below.",
                              uiOutput("dyn_type"),
                              uiOutput("dyn_area3"),
                              uiOutput("dyn_spp3"),
                              uiOutput("dyn_method3"),
                              uiOutput("dyn_watertype3"),
                              downloadButton("filterdownloaduu", "Download plot data")
                            ),
                            mainPanel(
                              tabsetPanel(
                                tabPanel("Instructions", 
                                         uiOutput("instructions"), 
                                         DTOutput("example")),
                                tabPanel("Comparisons", 
                                         br(), 
                                         h6("If no plots are displayed, upload a dataset to generate comparison plots."), 
                                         br(), 
                                         plotDownloadUI("LF_plot_UU"),
                                         p("Proportional size distribution length frequency. Black lines indicate standard error."),
                                         hr(),
                                         plotDownloadUI("RW_plot_UU"),
                                         p("Relative weight by proportional size distribution categories. Points indicate means and lines indicate standard error."),
                                         hr(),
                                         plotDownloadUI("CPUE_plot_UU", height = "200px"), 
                                         p("Catch per unit effort. The box represents the middle 50% of the standard data with the median value indicated by the line inside. The whiskers extend to the smallest and largest values within 1.5 times the inter quartile range and any individual points outside are outliers. The dashed line represents CPUE for the waterbody. CPUE units depend on collection method: boat and raft electrofishing are fish per hour, gill net is fish per net nights, drifting trammel net is fish per 100-m drift, large catfish hoopnet is fish per 24 hour set, bag seine is fish per 0.25 arc (small standing waters) or 0.5 arc (rivers), stream seine is fish per 10-15m haul, and backpack and tow barge electrofishing are fish per 100m²."))
                              )
                            )
                            
                          ), 
                          
                 ),
                 bslib::nav_item(tags$a(
                   tags$img(src='AFS_logo.png', style="background-color:white;"),
                   href = "https://fisheries.org/",
                   style='position:absolute;right:10px;bottom:5px;'
                 ))
                 
)




server <- function(input, output) {
  
  output$pics <- renderImage({
    list(
      src = file.path("www/fish_circles2.png"), 
      contentType = "image/png", 
      height = 365, 
      width = 819
    )
  }, deleteFile = FALSE)
  
  output$logo <- renderImage({
    list(
      src = file.path("www/Standardsamplingsponsors.png"), 
      contentType = "image/png", 
      height = 151, 
      width = 741
    )
  }, deleteFile = FALSE)
  
  # Render a UI for selecting area depending on North America, Ecoregions, or State/Province
  output$dyn_area <- renderUI({
    if(input$typechoice == "North America") {
      temp <- "North America"
    } else if(input$typechoice == "Ecoregion") {
      temp <- uni.area[1:12]
    } else if(input$typechoice == "State/Province") {
      temp2 <-  uni.area[uni.area %nin% 'North America']
      temp <- temp2[-1:-12]
    }
    
    selectInput(inputId = "areachoice",
                label = "Select area:", 
                choices = temp,
                selected = temp[1],
                multiple = TRUE)
  })
  
  # Render a UI for selecting area depending on North America, Ecoregions, or State/Province for tab2
  output$dyn_area2 <- renderUI({
    if(input$typechoice2 == "North America") {
      temp <- "North America"
    } else if(input$typechoice2 == "Ecoregion") {
      temp <- uni.area[1:12]
    } else if(input$typechoice2 == "State/Province") {
      temp2 <-  uni.area[uni.area %nin% 'North America']
      temp <- temp2[-1:-12]
    }
    
    selectInput(inputId = "areachoice2",
                label = "Select area:", 
                choices = temp,
                selected = temp[1])
  })
  
  # Read in raw user upload data to be used in reactive UI's
  uu_raw <- reactive({
    inFile <- input$upload
    uu <- read_csv(inFile$datapath)
    
    invalid_species_names <- uu %>% select(common_name) %>% filter(common_name %nin% unique(metrics$common_name)) %>% pull()
    
    if(any(is.na(uu))){
      validate("Uploaded dataset must have no empty rows")
    }
    
    # Return informative messages if uu data format is incorrect
    validate(
      need("state" %in% colnames(uu), "Uploaded dataset is missing state column"),
      need("waterbody_name" %in% colnames(uu), "Uploaded dataset is missing waterbody_name column"),
      need("common_name" %in% colnames(uu), "Uploaded dataset is missing common_name column"),
      need("method" %in% colnames(uu), "Uploaded dataset is missing method column"),
      need("year" %in% colnames(uu), "Uploaded dataset is missing year column"),
      need(n_distinct(uu$state) == 1, "State column should contain only one state"),
      need(n_distinct(uu$waterbody_name) == 1, "Waterbody column should contain only one name"), 
      need(length(invalid_species_names) == 0, paste0("Species name must match one in the provided list in instructions tab. These names are not valid: ", invalid_species_names))
    )
    
    # Duplicate records for all types
    uu_state <- uu %>% 
      select(-one_of("ecoregion")) %>% 
      mutate(type = "state") %>% 
      rename(area = state)
    
    if ("ecoregion" %in% names(uu)) {
      uu_ecoregion <- uu %>% 
        select(-state) %>% 
        mutate(type = "ecoregion") %>% 
        rename(area = ecoregion)
    }
    
    uu_all <- uu %>% 
      select(-state) %>% 
      select(-one_of("ecoregion")) %>% 
      mutate(type = "all", area = "North America")
    
    uu_all <- bind_rows(uu_all, uu_state)
    if(exists("uu_ecoregion")) uu_all <- bind_rows(uu_all, uu_ecoregion)
    
    if(exists("uu_state")){
      rm(uu_state)
    }
    
    if(exists("uu_ecoregion")){
      rm(uu_ecoregion)
    }
    
    print(uu_all)
    
  })
  
  # Render a UI for selecting area 
  output$dyn_type <- renderUI({
    req(input$upload)
    
    temp <- uu_raw() %>%
      select(type) %>%
      unique() %>%
      mutate(types = case_when(type == "all" ~ "North America",
                               type == "ecoregion" ~ "Ecoregion",
                               type == "state" ~ "State/Province")) %>%
      slice(match(c("North America", "Ecoregion", "State/Province"), types)) %>% 
      pull(types)
    
    radioGroupButtons(inputId = "typechoice3",
                      label = "Show data by:",
                      choices = temp,
                      direction = "vertical",
                      selected = temp[1])
  })
  
  # Render a UI for selecting area depending on North America, Ecoregions, or State/Province for tab 3
  output$dyn_area3 <- renderUI({
    req(input$upload)
    
    if(input$typechoice3 == "North America") {
      temp <- "North America"
    } else if(input$typechoice3 == "Ecoregion") {
      temp <- uu_raw() %>%
        filter(type == "ecoregion") %>%
        select(area) %>%
        unique() %>%
        pull()
    } else if(input$typechoice3 == "State/Province") {
      temp <- uu_raw() %>%
        filter(type == "state") %>%
        select(area) %>%
        unique() %>%
        pull()
    }
    
    selectInput(inputId = "areachoice3",
                label = "Select area:", 
                choices = temp,
                selected = temp[1])
  })  
  
  # Render a UI for selecting species depending on user upload data in tab 3
  output$dyn_spp3 <- renderUI({
    req(input$upload)
    req(input$areachoice3)
    
    temp <- uu_raw() %>%
      filter(area == input$areachoice3) %>%
      select(common_name) %>%
      unique() %>%
      pull()
    
    selectInput(inputId = "sppchoice3",
                label = "Select species:",
                choices = temp,
                selected = temp[1])
  })  
  
  # Render a UI for selecting species depending on user upload data in tab 3
  output$dyn_method3 <- renderUI({
    req(input$upload)
    req(input$areachoice3)
    req(input$sppchoice3)
    
    temp <- uu_raw() %>%
      filter(area == input$areachoice3,
             common_name == input$sppchoice3) %>%
      select(method) %>%
      unique() %>%
      pull()
    
    selectInput(inputId = "methodchoice3",
                label = "Select method:",
                choices = temp,
                selected = temp[1])
  })  
  
  # Render a UI for selecting species depending on user upload data in tab 3
  output$dyn_watertype3 <- renderUI({
    req(input$upload) 
    req(input$areachoice3)
    req(input$sppchoice3)
    req(input$methodchoice3)
    
    temp <- uu_raw() %>%
      filter(area == input$areachoice3,
             common_name == input$sppchoice3,
             method == input$methodchoice3) %>%
      select(waterbody_type) %>%
      unique() %>%
      pull()
    
    selectInput(inputId = "watertypechoice3",
                label = "Select waterbody type:",
                choices = temp,
                selected = temp[1])
  }) 
  
  
  # make filtered, a reactive data object for tab 1 explore, multiple combos okay
  filtered <- reactive({
    
    f_df <-  metrics %>%
      filter(metric %in% input$metricchoice,
             area %in% input$areachoice,
             case_when("All" %in% input$sppchoice ~ common_name %in% uni.spp,
                       "All" %nin% input$sppchoice ~ common_name %in% input$sppchoice),
             case_when("All" %in% input$methodchoice ~ method %in% uni.method,
                       "All" %nin% input$methodchoice ~ method %in% input$methodchoice),
             case_when("All" %in% input$watertypechoice ~ waterbody_type %in% uni.watertype,
                       "All" %nin% input$watertypechoice ~ waterbody_type %in% input$watertypechoice)) %>%
      arrange(common_name) %>%
      select(area, common_name, method, waterbody_type, gcat, metric, N, mean,
             se, `5%`, `25%`, `50%`, `75%`, `95%`)
    
  })
  
  # For downloading the entire dataset for tab 1 explore
  all <- reactive({
    
    all_df <-  metrics %>%
      arrange(common_name) %>%
      select(area, common_name, method, waterbody_type, gcat, metric, N, mean, 
             se, `5%`, `25%`, `50%`, `75%`, `95%`) 
  })
  
  output$alldownload <- downloadHandler(
    filename = function() {
      paste0("AFS-all-", Sys.Date(), ".csv")
    },
    
    content = function(file) {
      write.csv(all(), file)
    }
  )
  
  # make locations, a reactive data object for tab 1, only locations that are selected
  locate <- reactive({
    if(input$typechoice == "North America") {
      l_df <-  locs %>%
        filter(case_when("All" %in% input$sppchoice ~ common_name %in% uni.spp,
                         "All" %nin% input$sppchoice ~ common_name %in% input$sppchoice),
               case_when("All" %in% input$methodchoice ~ method %in% uni.method,
                         "All" %nin% input$methodchoice ~ method %in% input$methodchoice),
               case_when("All" %in% input$watertypechoice ~ waterbody_type %in% uni.watertype,
                         "All" %nin% input$watertypechoice ~ waterbody_type %in% input$watertypechoice)) %>%
        mutate(coords = paste(lat, long)) %>%
        group_by(state, ecoregion, waterbody_name, common_name, method, 
                 waterbody_type, coords) %>%
        summarize(Nsurveys = n(),
                  lat = unique(lat),
                  long = unique(long)) %>%
        ungroup()%>%
        select(-coords)
      
    } else if(input$typechoice == "Ecoregion") {
      l_df <-  locs %>%
        filter(ecoregion %in% input$areachoice,
               case_when("All" %in% input$sppchoice ~ common_name %in% uni.spp,
                         "All" %nin% input$sppchoice ~ common_name %in% input$sppchoice),
               case_when("All" %in% input$methodchoice ~ method %in% uni.method,
                         "All" %nin% input$methodchoice ~ method %in% input$methodchoice),
               case_when("All" %in% input$watertypechoice ~ waterbody_type %in% uni.watertype,
                         "All" %nin% input$watertypechoice ~ waterbody_type %in% input$watertypechoice)) %>%
        mutate(coords = paste(lat, long)) %>%
        group_by(state, ecoregion, waterbody_name, common_name, method, 
                 waterbody_type, coords) %>%
        summarize(Nsurveys = n(),
                  lat = unique(lat),
                  long = unique(long)) %>%
        ungroup()%>%
        select(-coords)
      
    } else if(input$typechoice == "State/Province") {
      l_df <-  locs %>%
        filter(state %in% input$areachoice,
               case_when("All" %in% input$sppchoice ~ common_name %in% uni.spp,
                         "All" %nin% input$sppchoice ~ common_name %in% input$sppchoice),
               case_when("All" %in% input$methodchoice ~ method %in% uni.method,
                         "All" %nin% input$methodchoice ~ method %in% input$methodchoice),
               case_when("All" %in% input$watertypechoice ~ waterbody_type %in% uni.watertype,
                         "All" %nin% input$watertypechoice ~ waterbody_type %in% input$watertypechoice)) %>%
        mutate(coords = paste(lat, long)) %>%
        group_by(state, ecoregion, waterbody_name, common_name, method, 
                 waterbody_type, coords) %>%
        summarize(Nsurveys = n(),
                  lat = unique(lat),
                  long = unique(long)) %>%
        ungroup() %>%
        select(-coords)
      
    }
  })
  
  # Create excluded dataframe for combinations with only one site and one year
  exclude <- reactive({
    if(input$typechoice == "North America") {
      e_df <- locate() %>%
        group_by(common_name, method, waterbody_type) %>%
        summarize(Nlocs = length(waterbody_name),
                  Nsurveys_tot = sum(Nsurveys))%>%
        filter(Nlocs == 1,
               Nsurveys_tot == 1) %>%
        select(-Nlocs, -Nsurveys_tot)
      
    } else if(input$typechoice == "Ecoregion") {
      e_df <- locate() %>%
        group_by(ecoregion, common_name, method, waterbody_type) %>%
        summarize(Nlocs = length(waterbody_name),
                  Nsurveys_tot = sum(Nsurveys))%>%
        filter(Nlocs == 1,
               Nsurveys_tot == 1) %>%
        select(-Nlocs, -Nsurveys_tot)
      
    } else if(input$typechoice == "State/Province") {
      e_df <- locate() %>%
        group_by(state, common_name, method, waterbody_type) %>%
        summarize(Nlocs = length(waterbody_name),
                  Nsurveys_tot = sum(Nsurveys)) %>%
        filter(Nlocs == 1,
               Nsurveys_tot == 1) %>%
        select(-Nlocs, -Nsurveys_tot)
      
    }
  })
  
  # make filtered, a reactive data object for tab 2, single combos only
  filtered2 <- reactive({
    
    f_df <-  metrics %>%
      filter(metric %in% uni.metric,
             area %in% input$areachoice2,
             common_name %in% input$sppchoice2,
             method %in% input$methodchoice2,
             waterbody_type %in% input$watertypechoice2) %>%
      arrange(common_name) %>%
      select(area, common_name, method, waterbody_type, gcat, metric, N, mean,
             se, `5%`, `25%`, `50%`, `75%`, `95%`)
    print(f_df)
    
  })
  
  observeEvent(input$areachoice2, {
    print(paste0("You have chosen: ", input$areachoice2))
  })
  
  output$filterdownload <- downloadHandler(
    filename = function() {
      paste0("AFS-filtered-", Sys.Date(), ".csv")
    },
    
    content = function(file) {
      write_csv(filtered(), file)
    }
  )
  
  # Making the table look nice
  output$filtertable <- renderDT({  
    
    filtered() %>%
      mutate(method = str_replace_all(method, "_", " "),
             area = str_replace_all(area, "_", " "),
             waterbody_type = str_replace_all(waterbody_type, "_", " ")) %>%
      dplyr::mutate(N = as.integer(N),
                    mean = round(mean, 2),
                    se = round(se, 2),
                    `5%` = round(`5%`, 2), 
                    `25%` = round(`25%`, 2),
                    `50%` = round(`50%`, 2),
                    `75%` = round(`75%`, 2),
                    `95%` = round(`95%`, 2)) %>%
      dplyr::rename("Area" = area,
                    "Common Name" = common_name,
                    "Method" = method,
                    "Waterbody Type" = waterbody_type,
                    "Proportional Size Distribution" = gcat,
                    "Metric" = metric, 
                    "N" = N,
                    "Mean" = mean,
                    "SE" = se,
                    "5%" = `5%`,
                    "25%" = `25%`,
                    "50%" = `50%`,
                    "75%" = `75%`,
                    "95%" = `95%`)
  })
  
  filtered3 <- reactive({
    
    f_df <-  metrics %>%
      filter(metric %in% uni.metric,
             area %in% input$areachoice3,
             common_name %in% input$sppchoice3,
             method %in% input$methodchoice3,
             waterbody_type %in% input$watertypechoice3) %>%
      arrange(common_name) %>%
      select(area, common_name, method, waterbody_type, gcat, metric, N, mean,
             se, `5%`, `25%`, `50%`, `75%`, `95%`)
    
  })
  
  uu_processed <- reactive({
    # Suppresses error messages for plots until inputs are chosen
    req(input$areachoice3)
    req(input$sppchoice3)
    req(input$methodchoice3)
    req(input$watertypechoice3)
    
    inFile <- input$upload
    uu <- read_csv(inFile$datapath)
    print("UU original")
    print(uu)
    
    # Duplicate records for all types
    uu_state <- uu %>% 
      select(-one_of("ecoregion")) %>% 
      mutate(type = "state") %>% 
      rename(area = state)
    
    if ("ecoregion" %in% names(uu)) {
      uu_ecoregion <- uu %>% 
        select(-state) %>% 
        mutate(type = "ecoregion") %>% 
        rename(area = ecoregion)
    }
    
    uu_all <- uu %>% 
      select(-state) %>% 
      select(-one_of("ecoregion")) %>% 
      mutate(type = "all", area = "North America")
    
    uu_all <- bind_rows(uu_all, uu_state)
    if(exists("uu_ecoregion")) uu_all <- bind_rows(uu_all, uu_ecoregion)
    
    # Calculate 3 metrics for user data
    uu_counts <- uu_all %>% 
      group_by(type, area, common_name, method, waterbody_type) %>%
      summarise(N = n())
    
    uu_cpue <- calculate_cpue(uu_all)
    uu_lf <- calculate_lf(uu_all)
    uu_rw <- calculate_rw(uu_all)
    
    uu_process <- bind_rows(uu_cpue, uu_lf, uu_rw) %>% 
      left_join(uu_counts, by = c("type", "area", "common_name", "method", "waterbody_type")) %>% 
      mutate(gcat = case_when(gcat == "stock" ~ "S-Q",
                              gcat == "quality" ~ "Q-P",
                              gcat == "preferred" ~ "P-M",
                              gcat == "memorable" ~ "M-T",
                              gcat == "trophy" ~ "T") %>%
               factor(levels = c("S-Q", "Q-P", "P-M", "M-T", "T")))
    
    f_df <-  uu_process %>%
      filter(metric %in% uni.metric,
             area %in% input$areachoice3,
             common_name %in% input$sppchoice3,
             method %in% input$methodchoice3,
             waterbody_type %in% input$watertypechoice3) %>%
      arrange(common_name) %>%
      select(area, common_name, method, waterbody_type, gcat, metric, N, mean,
             se, `5%`, `25%`, `50%`, `75%`, `95%`)
    
    print("UU filtered")
    print(f_df$metric)
    print(f_df, n = Inf)
    
  })
  
  # For downloading the filtered dataset for tab 3 comparison
  
  both <- reactive({
    
    both_df <- uu_processed() %>% 
      bind_rows(filtered3(), .id = "id") %>% 
      rename(data_source = id) %>% 
      mutate(data_source = case_when(data_source == 1 ~ "User upload", 
                                     data_source == 2 ~ "Standardized")) %>%
      arrange(metric) %>%
      select(metric, data_source, area, common_name, method, waterbody_type, gcat, N, mean, 
             se, `5%`, `25%`, `50%`, `75%`, `95%`) 
  })
  
  output$filterdownloaduu <- downloadHandler(
    filename = function() {
      paste0("AFS-UU-filtered-", Sys.Date(), ".csv")
    },
    
    content = function(file) {
      write.csv(both(), file)
    }
  )
  
  # Map of ecoregions and sites
  output$plotSites <- renderLeaflet({
    
    # Remove 'exclude()' from plot_data if only 1 site or 1 year is represented
    plot_data <- locate() %>%
      anti_join(exclude())
    
    # print(plot_data)
    factpal <- colorFactor(rainbow(length(unique(ecoregions_trans$NA_L1CODE))), 
                           ecoregions_trans$NA_L1CODE)
    
    map <- leaflet() %>% 
      addTiles(options = tileOptions(minZoom = 2, maxZoom = 6)) %>%
      addPolygons(data = ecoregions_trans, color = ~factpal(NA_L1CODE),
                  fillOpacity = 0.5, popup = ~htmltools::htmlEscape(NA_L1NAME),
                  stroke = FALSE)
    if(nrow(plot_data) > 0) {
      map <- map %>%
        addCircleMarkers(data = plot_data, lng = ~long, lat = ~lat, stroke = FALSE, 
                         radius = 3, fillOpacity = 0.75, fillColor = "black",
                         popup = popupTable(plot_data,
                                            zcol = 4:7,
                                            row.numbers = FALSE,
                                            feature.id = FALSE))
    }
    print(map)
    
  })
  
  # Text to describe insufficient sample size for anonymity
  output$report_absent1 <- renderText({
    paste0(nrow(exclude()), " site(s) not shown to maintain data anonymity. ")
  })
  
  # Text to describe sites missing lat/lon
  output$report_absent2 <- renderText({
    temp <- locate() %>% filter(is.na(lat)) 
    if(nrow(temp) == 0) {
      paste0(nrow(temp), 
             " site(s) not shown due to unreported coordinates. ")
    } else {
      paste0(nrow(temp), 
             " site(s) not shown due to unreported coordinates in ",
             paste(unique(temp$state), collapse = ", "), 
             ". ") 
    }
    
  })
  
  
  plotLengthFrequency <- reactive({
    temp <- filtered2() %>%
      filter(metric == "Length Frequency")
    
    N <- temp %>% 
      distinct(N) %>% 
      pull(N)
    
    if(nrow(temp) == 0){
      fig <- ggplot() +
        annotate("text", x = 1, y = 1, size = 8,
                 label = "No length frequency data for selected options") +
        theme_void()
    } else {
      fig <- ggplot(temp, aes(x = gcat, y = mean)) +
        geom_bar(stat = "identity", fill = "black") +
        geom_errorbar(aes(ymin = mean - se,
                          ymax = mean + se),
                      width = 0) +
        scale_y_continuous("Frequency (%)",
                           limits = c(0, 100),
                           expand = c(0, 0)) +
        theme_classic(base_size = 16) +
        xlab("Proportional size distribution categories") +
        theme(legend.position = "none") +
        geom_richtext(x = 5, y = 90, label = paste0("<i>N</i> = ", N), size = 7, fill = NA, label.color = NA)
    }
    
    print(fig)
    
  })
  
  callModule(plotDownload, "LF_plot", plotLengthFrequency)
  
  plotRelativeWeight <- reactive({
    
    temp <- filtered2() %>%
      filter(metric == "Relative Weight")
    
    if(nrow(temp) == 0){
      fig <- ggplot() +
        annotate("text", x = 1, y = 1, size = 8,
                 label = "No relative weight data for selected options") +
        theme_void()
    } else {
      fig <- ggplot(temp, aes(x = gcat)) +
        geom_point(aes(y = mean),
                   size = 4, 
                   color = "black") +
        geom_errorbar(aes(ymin = mean - se, ymax = mean + se), 
                      width = 0, 
                      color = "black") +
        geom_hline(yintercept = 100, linetype = "dashed") +
        scale_x_discrete(drop = FALSE) +
        ylim(50, 160) +
        labs(y = "Relative weight (<i>W<sub>r</sub></i>)") +
        theme_classic(base_size = 16) +
        xlab("Proportional size distribution categories") +
        theme(legend.position = "bottom",
              legend.title = element_blank(), 
              axis.title.y = ggtext::element_markdown())
    }    
    
    print(fig)
    
  })
  
  callModule(plotDownload, "RW_plot", plotRelativeWeight)
  
  
  plotCPUE <- reactive({
    
    temp <- filtered2() %>%
      filter(metric == "CPUE") 
    
    N <- unique(temp$N)
    
    if(nrow(temp) == 0){
      fig <- ggplot() +
        annotate("text", x = 1, y = 1, size = 8,
                 label = "No CPUE data for selected options") +
        theme_void()
    } else {
      fig <- ggplot(temp, aes(y = area)) +
        geom_boxplot(aes(xmin = `5%`,
                         xlower = `25%`,
                         xmiddle = `50%`,
                         xupper = `75%`,
                         xmax = `95%`),
                     stat = "identity", 
                     fill = "white") +
        scale_x_continuous("CPUE") +
        theme_classic(base_size = 16) +
        labs(title = paste0("*N* = ", N)) +
        theme(axis.title.y = element_blank(),
              axis.text.y = element_blank(),
              axis.ticks.y = element_blank(), 
              plot.title = ggtext::element_markdown(hjust = 1, face = "bold"))
    }
    
    print(fig)
    
  })
  
  callModule(plotDownload, "CPUE_plot", plotCPUE)
  
  mtext <- "
<br>
  
<center><b>How to Format Input Data</b></center>

You can upload your own data to compare to the standardized data. This needs to be provided as a csv file that will have length and/or weight measurements with one row per observation (a single fish or multiple fish summed). This app will calculate the three metrics of interest for each unique combination of area, species, collection method, type of water body, and year. The three metrics are: 

1. Catch per unit effort (CPUE)
2. Length frequency
3. Relative weight

<center><b>Column Details</b></center>

Below are the details of the required columns in the data. The order of the columns does not matter <b>BUT</b> the column names are <b>case sensitive and must be lower case</b>. Additionally, only upload data from <b>one</b> waterbody sampling effort at a time (under the column `waterbody_name`). There is an example dataset at the bottom of the page with simulated data that may be a helpful guide.

1. CPUE requires `effort` column
2. Length frequency requires `total_length` column 
3. Relative weight requires `weight` and `total_length` columns

Required columns in input dataframe: 

- **Location**: 
  - **Requires** `state` column
    - `state` is name of state where fish were collected, spelled out with first letter capitalized
    - Every row should be the same state (as it's for the same water body)
  - An `ecoregion` column can optionally be included, to compare your data to standardized data in that ecoregion
    - `ecoregion` is name of ecoregion where fish were collected, spelled out and capitalized correctly
    - You can also find your ecoregion by clicking on your location the map on the Explore tab. Ecoregions can also be determined from [EPA ecoregions](https://www.epa.gov/eco-research/ecoregions-north-america) level 1.  Correctly formatted ecoregion names are listed below. 
  - `waterbody_name` is the name of the water body

<style>
.basic-styling td,
.basic-styling th {
  border: 1px solid #999;
  padding: 0.5rem;
}
</style>

<div class='ox-hugo-table basic-styling'>
<div></div>
<div class='table-caption'>
  <span class='table-number'></span>
</div>

| **Ecoregions**                    |
|-----------------------------------|
| 0 Water                           |
| 1 Arctic Cordillera               |
| 2 Tundra                          |
| 3 Taiga                           |
| 4 Hudson Plain                    |
| 5 Northern Forests                |
| 6 Northwestern Forested Mountains |
| 7 Marine West Coast Forest        |
| 8 Eastern Temperate Forests       |
| 9 Great Plains                    |
| 10 North American Deserts         |
| 11 Mediterranean California       |
| 12 Southern Semiarid Highlands    |
| 13 Temperate Sierras              |
| 14 Tropical Dry Forests           |
| 15 Tropical Wet Forests           |

<br>

- **Date**: 
  - `year` is a four-digit numeric
- **Measurements**: 
  - `total_length` is fish record length (mm)
  - `weight` is fish record weight (g)
  - `effort` is specified in **Collection method**
- **Collection method**: see the table below for details
  - `method` must exactly match one of the options in 'Method name'
  - `effort` is the **total** effort of the survey, report number of effort units
      - Examples of effort: 4,556 seconds for electrofishing, 36 net nights for gill net surveys, 5 100-m drifts for drifting trammel net
  - The 'gill_net_spring' method is for gill netting done between January and June (months 1 - 6), and 'gill_net_fall' is for between July and December (months 7 - 12)
  - Data for additional collection methods will be added in the future

<center>

<style>
.basic-styling td,
.basic-styling th {
  border: 1px solid #999;
  padding: 0.5rem;
}
</style>

<div class='ox-hugo-table basic-styling'>
<div></div>
<div class='table-caption'>
  <span class='table-number'></span>
</div>

| **Method**          | **Effort Units** |
|--------------------------|-----------------|
| boat_electrofishing      | Seconds         |
| raft_electrofishing      | Seconds         |
| gill_net_fall            | Net nights      |
| gill_net_spring          | Net nights      |
| drifting_trammel_net     | 100-m drift     |
| large_catfish_hoopnet    | 24 hour set     |
| bag_seine                | 0.25 arc (small_standing_waters); 0.5 arc (rivers) |
| stream_seine             | 10-15 m haul    |
| backpack_electrofishing  | 100 m<sup>2</sup>            |
| tow_barge_electrofishing | 100 m<sup>2</sup>            |

</div>

</center>

- **Type of water body**:
  - `waterbody_type` must exactly match one of the following: *large_standing_waters*, *small_standing_waters*, *two_story_standing_waters*, *wadeable_streams*, *rivers*
- **Species**:
  - `common_name` must **exactly match** one of following species, as from [`FSA::PSDlit`](https://fishr-core-team.github.io/FSA/):

<center>

<style>
.basic-styling td,
.basic-styling th {
  border: 1px solid #999;
  padding: 0.5rem;
}
</style>

<div class='ox-hugo-table basic-styling'>
<div></div>
<div class='table-caption'>
  <span class='table-number'></span>
</div>

| **Species**        | **Species**      | **Species**               |
|--------------------|------------------|---------------------------|
| Arctic Grayling    | Gizzard Shad     | Shorthead Redhorse        |
| Bigmouth Buffalo   | Grass Carp       | Silver Carp               |
| Black Bullhead     | Green Sunfish    | Smallmouth Bass           |
| Black Crappie      | Lake Trout       | Spotted Bass              |
| Blue Catfish       | Largemouth Bass  | Spotted Gar               |
| Bluegill           | Longnose Gar     | Striped Bass (landlocked) |
| Brook Trout        | Muskellunge      | Utah Chub                 |
| Brown Bullhead     | Northern Pike    | Walleye                   |
| Brown Trout        | Paddlefish       | Warmouth                  |
| Bull Trout         | Palmetto Bass    | White Bass                |
| Burbot             | Pumpkinseed      | White Crappie             |
| Chain Pickerel     | Rainbow Trout    | White Perch               |
| Channel Catfish    | Redear Sunfish   | White Sucker              |
| Common Carp        | River Carpsucker | Yellow Bass               |
| Cutthroat Trout    | Rock Bass        | Yellow Bullhead           |
| Flathead Catfish   | Sauger           | Yellow Perch              |
| Freshwater Drum    | Saugeye          |                           |

</div>

</center>  

<br>

<center><b>Example Dataset</b></center>

<br>  
"

output$instructions <- renderUI({
  tf <- tempfile()
  knit(text = mtext, output = tf)
  HTML(markdown::markdownToHTML(file = tf))
})

output$example <- renderDT(datatable(ex, options = list(lengthChange = FALSE, 
                                                        pageLength = 25)) %>%
                             formatRound(c(8:10), 0, mark ="") %>%
                             formatString(7))

plotLengthFrequencyuser <- reactive({
  # uu_unfiltered() has a req() for uploaded data, so no more error messages
  temp <- uu_processed() %>% 
    bind_rows(filtered3(), .id = "id") %>% 
    rename(data_source = id) %>% 
    mutate(data_source = case_when(data_source == 1 ~ "User upload", 
                                   data_source == 2 ~ "Standardized")) %>% 
    filter(metric == "Length Frequency")
  print(temp)
  
  stand_N <- temp %>% 
    filter(data_source == "Standardized") %>% 
    distinct(N) %>% 
    pull(N)
  
  stand_N_label <- paste0("Standardized average <br> length frequency <br> (*N* = ", stand_N, " datasets)")
  
  user_N <- temp %>% 
    filter(data_source == "User upload") %>% 
    distinct(N) %>% 
    pull(N)
  
  user_N_label <- paste0("Waterbody length <br> frequency <br> (n = ", user_N, " fish)")
  
  stand_only <- temp %>% 
    filter(data_source == "Standardized")
  
  if(nrow(stand_only) == 0){
    fig <- ggplot() +
      annotate("text", x = 1, y = 1, size = 8,
               label = "No standardized data \n corresponding to uploaded data") +
      theme_void()
    
  } else {
    fig <- ggplot(temp, aes(x = gcat, y = mean, fill = data_source)) +
      geom_bar(stat = "identity", position = position_dodge(preserve = "single"), 
               color = "black")  +
      geom_errorbar(aes(ymin = mean - se,
                        ymax = mean + se),
                    position = position_dodge(width = 0.9, preserve = "single"),
                    width = 0) +
      scale_y_continuous("Frequency (%)",
                         limits = c(0, 100),
                         expand = c(0, 0)) +
      scale_fill_manual("legend", values = c("black", "grey60"), 
                        name = "Data source", 
                        labels = c(stand_N_label, user_N_label)) +
      theme_classic(base_size = 16) +
      xlab("Proportional size distribution categories") +
      theme(legend.position = c(0.85, 0.85), 
            legend.text = ggtext::element_markdown(), 
            legend.key.spacing.y = unit(0.25, "cm"))
  }
  
  print(fig)
  
})

callModule(plotDownload, "LF_plot_UU", plotLengthFrequencyuser)

plotRelativeWeightuser <- reactive({
  
  temp <- uu_processed() %>% 
    bind_rows(filtered3(), .id = "id") %>% 
    rename(data_source = id) %>% 
    mutate(data_source = case_when(data_source == 1 ~ "User upload", 
                                   data_source == 2 ~ "Standardized")) %>% 
    filter(metric == "Relative Weight")
  
  stand_only <- temp %>% 
    filter(data_source == "Standardized")
  
  if(nrow(stand_only) == 0){
    
    fig <- ggplot() +
      annotate("text", x = 1, y = 1, size = 8,
               label = "No standardized data \n corresponding to uploaded data") +
      theme_void()
    
  } else if (max(temp$mean) > 160){
    
    fig <- ggplot() +
      annotate("text", x = 1, y = 1, size = 8,
               label = "Relative weight exceeds max of 160") +
      theme_void()
    
  } else {
    
    fig <- ggplot(temp, aes(x = gcat)) +
      geom_point(aes(y = mean,
                     color = data_source),
                 size = 4, 
                 position = position_dodge(width = 0.05)) +
      geom_errorbar(aes(ymin = mean - se, 
                        ymax = mean + se, 
                        color = data_source),
                    width = 0, 
                    position = position_dodge(width = 0.05)) +
      scale_x_discrete(drop = FALSE) +
      ylim(50, 160) +
      geom_hline(yintercept = 100, linetype = "dashed") +
      labs(y = "Relative weight (<i>W<sub>r</sub></i>)") +
      scale_color_manual(name = "Data source", values = c("black", "grey60"), 
                         labels = c("Standardized data", "Waterbody")) +
      theme_classic(base_size = 16) +
      xlab("Proportional size distribution categories") +
      theme(legend.position = "inside", 
            legend.position.inside = c(0.85, 0.85), 
            axis.title.y = ggtext::element_markdown()) 
    
  }
  
  print(fig)
  
})

callModule(plotDownload, "RW_plot_UU", plotRelativeWeightuser)

plotCPUEuser <- reactive({
  temp <- uu_processed() %>% 
    bind_rows(filtered3(), .id = "id") %>% 
    rename(data_source = id) %>% 
    mutate(data_source = case_when(data_source == 1 ~ "User upload", 
                                   data_source == 2 ~ "Standardized")) %>% 
    filter(metric == "CPUE")
  print("Check")
  print(temp, n = Inf)
  
  stand_N <- temp %>% 
    filter(data_source == "Standardized") %>% 
    distinct(N) %>% 
    pull(N)
  
  stand_N_label <- paste0("Standardized data <br> (*N* = ", stand_N, " datasets)")
  
  user_N_label <- paste0("Waterbody")
  
  stand_only <- temp %>% 
    filter(data_source == "Standardized")
  
  value_uu_cpue <- temp %>% 
    filter(data_source == "User upload") %>% 
    select(mean) %>% 
    pull()
  
  value_uu_cpue_df <- data.frame(xintercept = value_uu_cpue)
  
  if(nrow(stand_only) == 0){
    
    fig <- ggplot() +
      annotate("text", x = 1, y = 1, size = 8,
               label = "No standardized data \n corresponding to uploaded data") +
      theme_void()
    
  } else {
    
    fig <- ggplot(temp, aes(y = area, fill = data_source)) +
      geom_boxplot(data = filter(temp, data_source == "Standardized"), 
                   aes(xmin = `5%`,
                       xlower = `25%`,
                       xmiddle = `50%`,
                       xupper = `75%`,
                       xmax = `95%`),
                   stat = "identity") +
      geom_vline(data = value_uu_cpue_df, aes(xintercept = xintercept, linetype = "Waterbody"), color = "grey40", size = 1.5) +
      scale_x_continuous("CPUE") +
      scale_fill_manual(name = "Data source", values = c("white", "grey60"), 
                        labels = c(stand_N_label, user_N_label)) +
      scale_linetype_manual("", values = c("Waterbody" = 2)) +
      theme_classic(base_size = 16) +
      theme(axis.title.y = element_blank(),
            axis.text.y = element_blank(),
            axis.ticks.y = element_blank(), 
            #legend.title = element_blank(), 
            legend.text = ggtext::element_markdown())
    
  }
  
  print(fig)
  
})

callModule(plotDownload, "CPUE_plot_UU", plotCPUEuser)

}

shinyApp(ui, server)