NEWS.md

In ijlyttle/colorpath: Build Color Paths

colorpath 0.0.1 (development version)

• Update to use newest colorio. (#100)

• Add functions to expand the domain of a palette: pth_palette_rescale_domain(), pth_rescaler_domain(). (#98)

• Add functions for designing categorical palettes:

• pth_color_sfclumsat(): to evaluate color from a surface, using luminance and saturation. (#94)

• pth_plot_lumchroma(): to view the separation of colors under color-vision-deficiency. (#96)

• Add functions to evaluate distance within categorical and quantitative palettes: pth_data_cat_euclid(), pth_data_cat_metric(), pth_data_qnt_euclid(), pth_data_qn_metric(). (#85)

• Modify the data frame returned pth_data_cvd() to contain a column hex_original, rather than index_color. (#82)

• Add function pth_mat_max_chroma() to generate max-chroma colors. Implemented for pth_surface. Also implement rbind() for pth_mat. (#80)

• Add function pth_colorspace_name() to get human-readable name for color space of a set of colors, a surface, or a palette. (#78)

• Add new way to specify surfaces: pth_new_surface(), along with pth_surface_data(), and a pth_surface generic for pth_plot_surface(). (#63, #75)

• Modify max-chroma calculation to avoid out-of-gamut errors. (#69)

• Add function check_colorio() to make sure that Python colorio meets minimum version. (#66)

• Reworked the philosophy of this package so that we can consider a class of color spaces that I call Lab-100-class. These color spaces:

  • are described using dimensions luminance, blue-yellow, green-red.
  • the value of black is c(0, 0, 0); the value of white is approximately c(100, 0, 0).
  • aspire to be perceptually uniform: the difference between colors scales with the Euclidean distance between colors.
  • have polar-coordinate analogues using hue and chroma.

  Such color spaces implemented here:

  • CIELAB
  • CIELUV
  • CAM02-UCS
  • CAM16-UCS
  • Jzazbz-100

  New color-space functions (#44):

  • pth_to_cartesian(), pth_to_polar(): to convert back-and-forth among coordinate systems. Note that pth_to_cartesian() has an argument chroma_min, used to preserve the hue where the chroma is zero. This means that values where chroma is at or near zero are imputed with an imperceptibly small chroma, allowing the hue to be recovered when pth_to_polar() is used.

  • pth_to_hex(), pth_new_hex(): Convert to hex codes, or designate a character vector as hex codes.

  • Convert to matrix using a color space, or designate that a matrix uses a color space:

    • pth_to_cielab(), pth_new_cielab(): CIELAB
    • pth_to_cieluv(), pth_new_cieluv(): CIELUV
    • pth_to_cam02ucs(), pth_new_cam02ucs(): CAM02-UCS
    • pth_to_cam16ucs(), pth_new_cam16ucs(): CAM16-UCS
    • pth_to_jzazbz100(), pth_new_jzazbz100(): Jzazbz, scaled to 100

  • whitepoints_cie1931(): Some of the color spaces need a reference white-point; this helper function provides the values. The default, "D65", seems to be used widely.

  The functions pth_to_hex(), pth_to_cieluv(), etc., form a "boat" of sorts. You can pipe values from one such function to another; you will always be describing the same colors. i.e. you will not "fall out of the boat". However, the functions pth_to_cartesian() and pth_to_polar() kick you "out of the boat". To get back "into the boat", you will need to use one of the pth_new_() functions.

  New gamut functions (#46):

  • pth_in_gamut(): Indicates if a color is inside (not outside) the RGB gamut.
  • pth_max_chroma(): Returns the maximum chroma for each color, expressed using its color space.
  • pth_clip_chroma(): Returns the colors, capping the chromas for out-of-gamut colors to their maximum chromas.

  New functions to support distance calculations (#48):

  • pth_distance_euclid(): Returns Euclidean distances between colors, given a color space.
  • pth_distance_metric(): Returns metric-based distances between colors.
  • pth_n_color(): Returns the number of colors in a vector or matrix.
  • pth_mat_replace_data(): Returns a pth_mat with new data.

  New functions to support palettes. A palette (function) returns a color for each given value between 0 and 1 (#50):

  • pth_new_hue_surface(): Returns a function (with S3 class pth_hue_surface) that, for each given luminance, returns a hue.

  • pth_new_chroma_trajectory(): Returns a function (with S3 class pth_chroma_trajectory) that, for each given x (between 0 and 1), returns a set of luminance and chroma; uses a Bézier curve under the hood.

  • pth_new_palette(): Returns a palette function (with S3 class pth_palette), given a hue-surface, chroma-trajectory, and a color space.

  • pth_new_palette_hex(): Returns a palette function, given a vector of hex-codes and an optional transformer-function.

  • pth_palette_rescale_reverse(): Returns a rescaled palette, following the same path but reversed.\

  • pth_palette_rescale_euclid(): Returns a rescaled palette, following the same path but stretched to become more perceptually-effective. Uses the Euclidean distance of a color space.

  • pth_palette_rescale_metric(): Returns a rescaled palette, as above, but uses a metric to determine distance.

  • pth_palette_join(): Returns a palette created by joining two palettes; useful for creating diverging scales.

  • pth_pal_input_discrete(): Returns a palette-like function that takes an integer as an input, returns that many colors. In other words, turn a continuous palette-function into a discrete palette-function.

  • pth_pal_output_hex(): Returns a palette-like function that returns a vector of hex-codes, rather than a matrix.

  • pth_data_surface_raster(): Returns a data frame to use for a raster plot of a hue surface.

  • pth_data_control_points(): Returns a data frame for the control points of a trajectory.

  • pth_data_palette(): Returns a data frame for the points in a palette, expressed using polar coordinates.

  • pth_plot_surface(): Returns a ggplot of a hue surface.

  • pth_layer_control_points(): Returns a ggplot layer for control points.

  • pth_layer_palette(): Returns a ggplot layer for palette points.

  • pth_plot_palette(): Returns a ggplot for a palette function.

New functions to support visualizing the sRGB gamut (also other colorsets) in polar coordinates, including color-vision deficiency (#59):

• pth_mat_gamut(): Returns representation of sRGB gamut in a given color space.

• pth_cvd_grid(), pth_cvd_grid_full(), pth_cvd_grid_none(): Helpers to return tibble for possibilities for color-vision deficiency and severity.

• pth_transformer(): Returns a function to transform to the colorspace of the object that generated it.

• pth_plot_polar(): Returns a ggplot of colors in the chroma-hue plane, using polar coordinates.

• Added function get_distance() to calculate perceptual distances on a palette-function. (#38)

• Added functions data_surface_hl(), plot_surface_hl() that work with hue-luminance surface functions. (#31)

• Added functions surface_hl() and df_hcl() to facilitate creating sets of consistent multi-hue palettes. (#29)

• Added functions pal_luv_rescale_x() and pal_luv_rescale_lum() to avoid creating rescaler functions explicitly. (#27)

• Added pkgdown site. (#22)

• Added print() method for cpath_pal_luv.

• Added function plot_cl() to plot chroma-luminance plane for a palette. (#15)

• Added argument chroma_min (default 0.01) to as_mat_luv() to keep meaningful hue at low chroma. (#15)

• Added function data_hcl() to get HCL data to visualize palettes. (#15)

• Refactored to add constructor for LUV palettes, store mat_luv control-points as attribute. (#19)

• Refactored (#1, continued):

  • added arguments rescale_path and n to pal_luv_bezier(), rescales the palette to be perceptually uniform (in LUV space) with respect to the input.
  • made rescaler_bezier() an internal function.
  • renamed function palette_bezier() to pal_luv_bezier(); added argument rescale_path, defaults TRUE, to indicate to rescale the input according to the path length in LUV space.
  • renamed function rescale_palette() to rescale_pal_luv().
  • renamed function rescaler_linear_luminance() to rescaler_lum().
  • renamed function rescaler_linear_input() to rescaler_x().
  • renamed S3 class cpath_palette_luv to cpath_pal_luv.
  • added function as_pal_disc() to coerce to a discrete-palette function.
  • renamed palette_hex() to as_pal_hex().
  • renamed function luv() to as_mat_luv().
  • renamed argument luv to mat_luv.

• Added function to convert LUV palette-function to hex-code palette-function, palette_hex(). (#14)

• Added function to rescale a palette function, rescale_palette(). (#10)

• Added rescaler function rescaler_linear_luminance(). (#11)

• Added datasets df_hcl_blues and mat_luv_blues, which describe the same example colorpath. (#8)

• Added rescaler functions: rescaler_linear_input() and rescaler_bezier(). (#6, #11)

• Added function palette_bezier() to create a palette function using a Bézier spline. (#4)

• Added function luv() to convert HCL data frame to LUV matrix. (#2)

• Added a NEWS.md file to track changes to the package.

ijlyttle/colorpath documentation built on Sept. 13, 2022, 7:44 p.m.