R/gt_plt_percentile_dot.R

In gtExtras: Extending 'gt' for Beautiful HTML Tables

#' Create a dot plot from 0 to 100
#' @param data The single value that will be used to plot the point.
#' @param palette A length 3 palette, used to highlight high/med/low
#' @param add_label A logical indicating whether to add the label or note. This will only be added if it is the first or last row.
#' @param width A numeric indicating the
#' @return gt table
#'
add_pcttile_plot <- function(data, palette, add_label, width) {
  if (data %in% c("NA", "NULL", NA, NULL)) {
    return("<div></div>")
  }
  stopifnot("Values must be between 0 and 100" = dplyr::between(data, 0, 100))

  df_in <- dplyr::tibble(
    x = data,
    y = 1,
    color = palette
  )

  out_pct_plt <- ggplot(df_in) +
    geom_vline(xintercept = 50, color = "black", linewidth = 0.5) +
    geom_vline(
      xintercept = c(0, 25, 75, 100),
      color = "grey",
      linewidth = 0.25
    ) +
    geom_hline(
      yintercept = 1,
      color = "lightgrey",
      linewidth = 0.25,
      linetype = "dotted"
    ) +
    geom_point(
      aes(x = .data$x, y = .data$y, fill = I(.data$color)),
      color = "black",
      size = 3,
      stroke = 0.5,
      shape = 21
    ) +
    theme_void() +
    coord_cartesian(
      xlim = c(0, 100),
      ylim = c(0.6, 1.2),
      clip = "off"
    )

  if (isTRUE(add_label)) {
    out_pct_plt <- out_pct_plt +
      geom_text(
        data = NULL,
        aes(x = 1, y = .61, label = "0"),
        hjust = 0,
        vjust = 0,
        size = 1.5,
        family = "mono",
        color = "black"
      ) +
      geom_text(
        aes(x = 99, y = 0.61, label = "100"),
        hjust = 1,
        vjust = 0,
        size = 1.5,
        family = "mono",
        color = "black"
      ) +
      geom_text(
        aes(x = 49, y = 0.61, label = "5"),
        hjust = 1,
        vjust = 0,
        size = 1.5,
        family = "mono",
        color = "black"
      ) +
      geom_text(
        aes(x = 51, y = 0.61, label = "0"),
        hjust = 0,
        vjust = 0,
        size = 1.5,
        family = "mono",
        color = "black"
      )
  } else {
    out_pct_plt <- out_pct_plt
  }

  out_name <- file.path(tempfile(
    pattern = "file",
    tmpdir = tempdir(),
    fileext = ".svg"
  ))

  ggsave(
    out_name,
    out_pct_plt,
    height = 5,
    width = width,
    dpi = 25.4,
    units = "mm",
    device = "svg"
  )

  img_plot <- readLines(out_name) %>%
    paste0(collapse = "") %>%
    gt::html()

  on.exit(file.remove(out_name), add = TRUE)

  img_plot
}

#' Create a dot plot for percentiles
#' @description Creates a percentile dot plot in each row. Can be used as an
#' alternative for a 0 to 100% bar plot. Allows for scaling values as well and
#' accepts a vector of colors for the range of values.
#' @param gt_object An existing gt table
#' @param column The column to transform to the percentile dot plot. Accepts `tidyeval`. All values must be end up being between 0 and 100.
#' @param palette A vector of strings of length 3. Defaults to `c('blue', 'lightgrey', 'red')` as hex so `c("#007ad6", "#f0f0f0", "#f72e2e")`
#' @param width A numeric, indicating the width of the plot in `mm`, defaults to 25
#' @param scale A number to multiply/scale the values in the column by. Defaults to 1, but can also be 100 if you have decimals.
#' @return a gt table
#' @export
#'
#' @section Examples:
#' ```r
#' library(gt)
#' dot_plt <- dplyr::tibble(x = c(seq(10, 90, length.out = 5))) %>%
#'   gt() %>%
#'   gt_duplicate_column(x,dupe_name = "dot_plot") %>%
#'   gt_plt_percentile(dot_plot)
#' ```
#' @section Figures:
#' \if{html}{\figure{gt_plt_percentile.png}{options: width=500}}
#'
#' @family Plotting
#' @section Function ID:
#' 3-8
gt_plt_percentile <- function(
  gt_object,
  column,
  palette = c("#007ad6", "#f0f0f0", "#f72e2e"),
  width = 25,
  scale = 1
) {
  gt_object %>%
    text_transform(
      locations = cells_body({{ column }}),
      fn = function(x) {
        x <- as.double(x) * scale
        n_vals <- 1:length(x)

        stopifnot(
          "Values must be scaled between 0 and 100" = dplyr::between(x, 0, 100)
        )

        col_pal <- scales::col_quantile(
          palette = palette,
          domain = c(0:100),
          reverse = TRUE,
          alpha = TRUE,
          n = 5
        )(x)

        add_label <- n_vals %in% c(min(n_vals), max(n_vals))

        mapply(add_pcttile_plot, x, col_pal, add_label, width, SIMPLIFY = FALSE)
      }
    )
}