nmode's Git Repositories - Rnaught/blob - inst/app/scripts/data.R

   1 # Main logic block for data-related interactions.
   2 data_logic <- function(input, output, react_values) {
   3   # Initialize a data frame to hold the datasets.
   4   react_values$data_table <- data.frame(
   5     Name = character(0),
   6     `Time units` = character(0),
   7     `Case counts` = character(0),
   8     check.names = FALSE
   9   )
  10
  11   manual_entry(input, output, react_values)
  12   upload_data(input, output, react_values)
  13   load_samples(input, output, react_values)
  14   render_data_table(output, react_values)
  15   render_plot(input, output, react_values, "Days")
  16   render_plot(input, output, react_values, "Weeks")
  17   delete_data(input, react_values)
  18   export_data(output, react_values)
  19 }
  20
  21 # Convert the input case counts string to an integer vector.
  22 tokenize_counts <- function(counts_str) {
  23   suppressWarnings(as.integer(unlist(strsplit(trimws(counts_str), ","))))
  24 }
  25
  26 # Render the plots for daily and weekly data when the data table is updated.
  27 render_plot <- function(input, output, react_values, time_units) {
  28   observe({
  29     datasets <- react_values$data_table[
  30       which(react_values$data_table[["Time units"]] == time_units),
  31     ]
  32
  33     data_plot <- plotly::plot_ly(type = "scatter", mode = "lines")
  34     if (nrow(datasets) > 0) {
  35       for (i in seq_len(nrow(datasets))) {
  36         counts <- tokenize_counts(datasets[i, 3])
  37         data_plot <- plotly::add_trace(data_plot,
  38           x = seq_along(counts) - 1, y = counts, name = datasets[i, 1]
  39         )
  40       }
  41     }
  42
  43     plot_title <- paste(
  44       if (time_units == "Days") "Daily" else "Weekly", "case counts"
  45     )
  46
  47     data_plot <- plotly::layout(data_plot, title = plot_title,
  48       xaxis = list(title = time_units), yaxis = list(title = "Cases")
  49     )
  50
  51     data_plot <- plotly::config(data_plot, displaylogo = FALSE,
  52       toImageButtonOptions = list(
  53         filename = paste0("Rnaught_data_", tolower(time_units), "_plot")
  54       )
  55     )
  56
  57     output[[paste0("data_plot_", tolower(time_units))]] <-
  58       plotly::renderPlotly(data_plot)
  59   })
  60 }
  61
  62 # Validate and add manually-entered datasets.
  63 manual_entry <- function(input, output, react_values) {
  64   observeEvent(input$data_bulk, {
  65     validate_data(input, output, react_values, "data_area")
  66   })
  67 }
  68
  69 # Validate and add datasets from a CSV file.
  70 upload_data <- function(input, output, react_values) {
  71   observeEvent(input$data_upload, {
  72     validate_data(input, output, react_values, "data_upload")
  73   })
  74 }
  75
  76 # Validate datasets and update the data table.
  77 validate_data <- function(input, output, react_values, data_source) {
  78   tryCatch(
  79     {
  80       if (data_source == "data_area") {
  81         datasets <- read.csv(text = input$data_area, header = FALSE, sep = ",")
  82       } else if (data_source == "data_upload") {
  83         datasets <- read.csv(
  84           file = input$data_upload$datapath, header = FALSE, sep = ","
  85         )
  86       }
  87
  88       names <- trimws(datasets[, 1])
  89       units <- trimws(datasets[, 2])
  90       counts <- apply(data.frame(datasets[, 3:ncol(datasets)]), 1,
  91         function(row) {
  92           row <- suppressWarnings(as.integer(row))
  93           toString(row[!is.na(row) & row >= 0])
  94         }
  95       )
  96
  97       warning_text <- ""
  98
  99       # Ensure the dataset names are neither blank nor duplicates.
 100       if (anyNA(names) || any(names == "")) {
 101         warning_text <- paste0(warning_text,
 102           "Each row must begin with a non-blank dataset name.<br>"
 103         )
 104       } else {
 105         if (length(unique(names)) != length(names)) {
 106           warning_text <- paste0(warning_text,
 107             "The rows contain duplicate dataset names.<br>"
 108           )
 109         }
 110         if (any(names %in% react_values$data_table[, 1])) {
 111           warning_text <- paste0(warning_text,
 112             "The rows contain dataset names which already exist.<br>"
 113           )
 114         }
 115       }
 116
 117       # Ensure the second entry in each row is a time unit equal to
 118       # "Days" or "Weeks".
 119       if (!all(units %in% c("Days", "Weeks"))) {
 120         warning_text <- paste0(warning_text,
 121           "The second entry in each row must be either 'Days' or 'Weeks'.<br>"
 122         )
 123       }
 124
 125       # Ensure the counts in each row have at least one non-negative integer.
 126       if (any(counts == "")) {
 127         warning_text <- paste0(warning_text,
 128           "Each row must contain at least one non-negative integer.<br>"
 129         )
 130       }
 131
 132       output[[paste0(data_source, "_warn")]] <- renderUI(HTML(warning_text))
 133
 134       if (warning_text == "") {
 135         # Add the new datasets to the data table.
 136         new_rows <- data.frame(names, units, counts)
 137         colnames(new_rows) <- c("Name", "Time units", "Case counts")
 138         react_values$data_table <- rbind(react_values$data_table, new_rows)
 139
 140         # Evaluate all existing estimators on the new datasets and update the
 141         # corresponding columns in the estimates table.
 142         update_estimates_cols(new_rows, react_values)
 143
 144         showNotification("Datasets added successfully.", duration = 3)
 145       }
 146     },
 147     error = function(e) {
 148       output[[paste0(data_source, "_warn")]] <- renderText(
 149         "The input does not match the required format."
 150       )
 151     }
 152   )
 153 }
 154
 155 # Load sample datasets.
 156 load_samples <- function(input, output, react_values) {
 157   observeEvent(input$data_samples, {
 158     names <- c()
 159     units <- c()
 160     counts <- c()
 161
 162     # COVID-19 Canada, March 2020 (weekly).
 163     if (input$covid_canada) {
 164       names <- c(names, "COVID-19 Canada 2020/03/03 - 2020/03/31")
 165       units <- c(units, "Weeks")
 166       counts <- c(counts, toString(Rnaught::COVIDCanada[seq(41, 69, 7), 2]))
 167     }
 168     # COVID-19 Ontario, March 2020 (weekly).
 169     if (input$covid_ontario) {
 170       names <- c(names, "COVID-19 Ontario 2020/03/03 - 2020/03/31")
 171       units <- c(units, "Weeks")
 172       counts <- c(counts,
 173         toString(Rnaught::COVIDCanadaPT[seq(10176, 10204, 7), 3])
 174       )
 175     }
 176
 177     if (length(names) == 0) {
 178       output$data_samples_warn <- renderText(
 179         "At least one sample dataset must be selected."
 180       )
 181     } else if (any(names %in% react_values$data_table[, 1])) {
 182       output$data_samples_warn <- renderText(
 183         "At least one of the selected dataset names already exist."
 184       )
 185     } else {
 186       output$data_samples_warn <- renderText("")
 187
 188       new_rows <- data.frame(names, units, counts)
 189       colnames(new_rows) <- c("Name", "Time units", "Case counts")
 190       react_values$data_table <- rbind(react_values$data_table, new_rows)
 191
 192       # Evaluate all existing estimators on the sample datasets and update the
 193       # corresponding columns in the estimates table.
 194       update_estimates_cols(new_rows, react_values)
 195
 196       showNotification("Datasets added successfully.", duration = 3)
 197     }
 198   })
 199 }
 200
 201 # Render the data table when new datasets are added.
 202 render_data_table <- function(output, react_values) {
 203   observe({
 204     output$data_table <- DT::renderDataTable(
 205       react_values$data_table, rownames = FALSE
 206     )
 207   })
 208 }
 209
 210 # Delete rows in the data table and the corresponding columns in the estimates
 211 # table.
 212 delete_data <- function(input, react_values) {
 213   observeEvent(input$data_delete, {
 214     rows_selected <- input$data_table_rows_selected
 215     react_values$data_table <- react_values$data_table[-rows_selected, ]
 216     react_values$estimates_table <-
 217       react_values$estimates_table[, -(rows_selected + 2)]
 218   })
 219 }
 220
 221 # Export data table as a CSV file.
 222 export_data <- function(output, react_values) {
 223   output$data_export <- downloadHandler(
 224     filename = function() {
 225       paste0("Rnaught_data_", format(Sys.time(), "%y-%m-%d_%H-%M-%S"), ".csv")
 226     },
 227     content = function(file) {
 228       write.csv(react_values$data_table, file, row.names = FALSE)
 229     }
 230   )
 231 }
 232
 233 # When new datasets are added, evaluate all existing estimators on them and
 234 # add new columns to the estimates table.
 235 update_estimates_cols <- function(datasets, react_values) {
 236   new_cols <- data.frame(
 237     matrix(nrow = nrow(react_values$estimates_table), ncol = nrow(datasets))
 238   )
 239   colnames(new_cols) <- datasets[, 1]
 240
 241   if (nrow(new_cols) > 0) {
 242     for (row in seq_len(nrow(new_cols))) {
 243       estimator <- react_values$estimators[[row]]
 244       for (col in seq_len(ncol(new_cols))) {
 245         new_cols[row, col] <- eval_estimator(estimator, datasets[col, ])
 246       }
 247     }
 248   }
 249
 250   react_values$estimates_table <- cbind(
 251     react_values$estimates_table, new_cols
 252   )
 253 }