R/palettes.R

In BrookingsInstitution/ggbrookings: Grammar of Graphics in Brookings Style

#' Brookings Institution palettes
#'
#' Vectors with hex-color codes that correspond to the color palettes outlined in the Brookings Style Guide.
#' x
#'
#' @source Brookings Style Guide
#' x
#'
#'
#' @family brookings palettes
#' @rdname brookings_palettes
#' @export
brookings_colors <-
  c(
    `brookings blue` = "#003A70",
    # Brand blue
    brand_blue_90 = "#022A4E",
    brand_blue_80 = "#003A70",
    brand_blue_70 = "#1A4E80",
    brand_blue_60 = "#326295",
    brand_blue_50 = "#517EAD",
    brand_blue_40 = "#7098C3",
    brand_blue_30 = "#8DADD0",
    brand_blue_20 = "#A8BDD5",
    brand_blue_10 = "#DDE5ED",

    # Vivid blue
    vivid_blue_90 = "#023147",
    vivid_blue_80 = "#004B6E",
    vivid_blue_70 = "#0061A0",
    vivid_blue_60 = "#1479BB",
    vivid_blue_50 = "#1E8AD6",
    vivid_blue_40 = "#3398EA",
    vivid_blue_30 = "#5AADF6",
    vivid_blue_20 = "#8BB8E8",
    vivid_blue_10 = "#BFDFFC",

    # Teal
    teal_90 = "#032B30",
    teal_80 = "#09484F",
    teal_70 = "#116470",
    teal_60 = "#1C8090",
    teal_50 = "#2A9AAD",
    teal_40 = "#3EB2C6",
    teal_30 = "#59C6DA",
    teal_20 = "#7CD9EA",
    teal_10 = "#A6E9F5",

    # Green
    green_90 = "#1A3404",
    green_80 = "#294D0A",
    green_70 = "#33660F",
    green_60 = "#45821B",
    green_50 = "#5CA632",
    green_40 = "#7DBF52",
    green_30 = "#9CD674",
    green_20 = "#BDED9D",
    green_10 = "#DEF5CC",

    # Yellow
    yellow_90 = "#594C09",
    yellow_80 = "#877414",
    yellow_70 = "#C7A70A",
    yellow_60 = "#E0BB00",
    yellow_50 = "#F2CD00",
    yellow_40 = "#FFDD00",
    yellow_30 = "#FFE926",
    yellow_20 = "#FFF170",
    yellow_10 = "#FFF9C2",

    # Orange
    orange_90 = "#663205",
    orange_80 = "#994B08",
    orange_70 = "#B85B0A",
    orange_60 = "#EF6A00",
    orange_50 = "#FF851A",
    orange_40 = "#FF9E1B",
    orange_30 = "#FFB24D",
    orange_20 = "#FEC87F",
    orange_10 = "#FBD9A5",

    # Red
    red_90 = "#660507",
    red_80 = "#A00D11",
    red_70 = "#CD1A1C",
    red_60 = "#E22827",
    red_50 = "#ED3A35",
    red_40 = "#F75C57",
    red_30 = "#F98B83",
    red_20 = "#FCB0AA",
    red_10 = "#FDD7D4",

    # Magenta
    magenta_90 = "#510831",
    magenta_80 = "#8D1655",
    magenta_70 = "#A82168",
    magenta_60 = "#BF317B",
    magenta_50 = "#D2468E",
    magenta_40 = "#E160A2",
    magenta_30 = "#EC81B7",
    magenta_20 = "#F5A8CF",
    magenta_10 = "#FAD4E7",

    # Purple
    purple_90 = "#3E2C72",
    purple_80 = "#533C91",
    purple_70 = "#6A50AD",
    purple_60 = "#7C60BF",
    purple_50 = "#8E72D0",
    purple_40 = "#9C82D9",
    purple_30 = "#B59DEA",
    purple_20 = "#D0BEF5",
    purple_10 = "#E9E0FC",

    # Grey
    `PMS 427 C` = "#B1B3b3",

    # The Hamilton Project colors
    THP_turquoise ='#007363',
      THP_ltgreen ='#A5D867',
      THP_dkblue ='#006983',
      THP_dkgreen ='#427730',
      THP_purple ='#6E2585',
      THP_grey ='#FC7F92',
      THP_ltblue ='#00ADD0',
      THP_green ='#69BE28',
      THP_orange ='#FF6E00',
      THP_yellow ='#FFB612',
      THP_ltpurple = '#C957BC'
  )

#' Function to extract brookings colors as hex codes
#'
#' @param ... Character names of Brookings_colors
#'
brookings_cols <- function(...) {
  cols <- c(...)

  if (is.null(cols)) {
    return(brookings_colors)
  }

  brookings_colors[cols]
}

#' @rdname brookings_palettes
#'
brookings_palettes <- list(
  # BRAND COLORS
  # Use colors from the brand palette whenever possible.
  brand1 = brookings_cols("brookings blue", "vivid_blue_20", "orange_40"),
  brand2 = brookings_cols("brookings blue", "orange_40", "PMS 427 C"),
  # Analogous
  # Different shades of the same hue, or of similar hues can be used when the associated
  # values are related
  analogous1 = brookings_cols("brookings blue", "vivid_blue_20"),
  analogous2 = brookings_cols("brookings blue", "teal_40"),

  # CONTRASTING
  # Colors on the opposite ends of the spectrum. Use Brookings Blue with Secondary colors.
  contrasting1 = brookings_cols("brookings blue", "orange_40"),
  contrasting2 = brookings_cols("brookings blue", "yellow_50"),

  # SEMANTIC
  # Where applicable, use colors that are associated with certain concepts. For e.g., option 1 shows male and female data, and option 2 show pros and cons or positive and negative data.
  semantic1 = brookings_cols("teal_30", "red_40"),
  semantic2 = brookings_cols("teal_60", "red_80", "yellow_60"),
  semantic3 = brookings_cols("teal_30", "red_40", "yellow_40"),
  pos_neg1 = brookings_cols("green_50", "red_70"),
  pos_neg2 = brookings_cols("green_50", "yellow_50", "red_70"),


  # POLITICAL
  # Use red and blue
  # of similar intensity to represent data related to political parties in the US.
  # Do not use the Brand Blue color family to represent a political party.
  political1 = brookings_cols("vivid_blue_60", "red_50"),
  political2 = brookings_cols("vivid_blue_30", "red_30"),
  political3 = brookings_cols("vivid_blue_60", "red_50", "yellow_60"),
  political4 = brookings_cols("vivid_blue_30", "red_30", "yellow_30"),

  # CATEGORICAL
  categorical = c("#2599adff", "#00649fff", "#fd9d1fff", "#f5cc05ff", "#de60a1ff", "#9e0d12ff"),
  categorical_expanded = c("#2A9AAD", "#00649F", "#FF9E1B", "#533C91", "#E160A2", "#A00D11", "#5CA632"),
  yor = c("#FF9E1B", "#FFF170"),

  # SEQUENTIAL
  sequential1 = c(
    "#00649fff", "#0f78baff", "#1c8ad6ff", "#2e97eaff", "#56adf6ff",
    "#87c4feff", "#bcdefbff"
  ),
  sequential2 = c("#0d636fff", "#008080ff", "#009a80ff", "#2bb275ff", "#6dc960ff", "#b1dc44ff", "#fce829ff"),

  # Diverging
  diverging = c("#0f78baff", "#739fceff", "#b1c5deff", "#efefefff", "#f6b5a9ff", "#f07867ff", "#e02928ff"),

  # MISCELLANEOUS
  # A pleasing option using Brookings Blue and accent yellow
  misc = brookings_cols("brookings blue", "teal_40", "yellow_50"),

  # THP Palettes
  thp_categorical1 = brookings_cols("THP_dkgreen","THP_green","THP_purple"),

  thp_categorical2 = brookings_cols("THP_dkblue","THP_green", "THP_ltblue", "THP_yellow", "THP_purple", "THP_orange"),

  thp_continuous_purple = brookings_cols("THP_ltpurple", "THP_purple"),

  thp_gender=brookings_cols("THP_dkblue", "THP_purple", "THP_ltblue", "THP_ltpurple")
)

#' Return function to interpolate a brookings color palette
#'
#' @param palette Character name of palette in brookings_palettes
#' @param reverse Boolean indicating whether the palette should be reversed
#' @param ... Additional arguments to pass to colorRampPalette()
#'
brookings_pal <- function(palette = "brand1", reverse = FALSE, ...) {
  pal <- brookings_palettes[[palette]]

  if (reverse) pal <- rev(pal)

  colorRampPalette(pal, ...)
}

#' Color scales in the Brookings style

#' @param palette Character name of brookings_palettes
#' @param discrete Boolean indicating whether color aesthetic is discrete or not
#' @param reverse Boolean indicating whether the palette should be reversed
#' @param ... Additional arguments passed to discrete_scale() or scale_color_gradientn(), used respectively when discrete is TRUE or FALSE.
#'
#' @section Palettes:
#' \describe{
#'   \item{Analogous}{Different shades of the same hue, or of similar hues can be used when the associated values are related.}
#'   \item{Contrasting}{Colors on the opposite ends of the spectrum. Use Brookings Blue with Secondary colors.}
#'   \item{Semantic}{Where applicable, use colors that are associated with certain concepts. For e.g., semantic1, semantic2, and semantic3 could show subsets of gender data (female, male and other).}
#'   \item{Positive & Negative}{Shows pros, cons and neutral, or positive, negative and neutral data.}
#'   \item{Political}{Use red and blue of similar intensity to represent data related to political parties in the US. Yellow in political3 and political4 represents ‘Independent’ category}
#'   \item{Categorical}{Use categorical palettes to distinguish discrete categories of data that do not have an inherent ordering.}
#'   \item{Sequential}{Sequential palettes can be used to show an inherent order or variations in numeric values.}
#'   \item{Diverging}{Diverging palettes are useful when dealing with negative and positive values or a range of values that have two extremes with a baseline central value, like zero. The Brookings diverging palette should uses two distinct hues of similar brightness and saturation with a neutral color in the middle. Using a discrete set of colors with evenly distributed gradation can improve clarity of values relative to a continuous palette.}
#'   \item{Misc}{A pleasing option using Brookings Blue and accent yellow.}
#' }
#'
#' @name scales_brookings
NULL

#' @rdname scales_brookings
#' @export
scale_color_brookings <- function(palette = "brand1", discrete = TRUE, reverse = FALSE, ...) {
  pal <- brookings_pal(palette = palette, reverse = reverse)

  if (discrete) {
    ggplot2::discrete_scale("colour", paste0("brookings_", palette), palette = pal, ...)
  } else {
    ggplot2::scale_color_gradientn(colours = pal(256), ...)
  }
}

#' @rdname scales_brookings
#' @export
scale_fill_brookings <- function(palette = "brand1", discrete = TRUE, reverse = FALSE, ...) {

  pal <- brookings_pal(palette = palette, reverse = reverse)

  if (discrete) {
    ggplot2::discrete_scale("fill", paste0("brookings_", palette), palette = pal, ...)
  } else {
    scale_fill_gradientn(colours = pal(256), ...)
  }
}