Changes in version 0.4.0                        

Breaking changes

  - Removed run_palette_lab() and the bundled Shiny app. The Palette Lab
    now lives in its own repository
    (https://github.com/viniciusoike/ekioplot-palette-lab) and runs
    entirely in the browser at
    https://viniciusoike.github.io/ekioplot-palette-lab/. This drops 8
    app-only Suggests (bslib, colorspace, colourpicker, forcats, ggbump,
    patchwork, shiny, stringr), leaving ekioplot a lean visualization
    package.

                        Changes in version 0.3.1                        

Documentation

  - Rewrote README.md to match the current API (the previous version
    referenced removed font helpers, theme style names, and palette
    names that no longer exist).

Infrastructure

  - Added an R-CMD-check GitHub Actions workflow covering Windows,
    macOS, and Linux (release, devel, and oldrel).
  - Removed stale repository artifacts and scratch files.

                        Changes in version 0.3.0                        

Breaking changes

  - Complete API overhaul. Palette system, themes, and scales have been
    redesigned with a cleaner, more consistent interface.

Color system

  - Unified palette access through ekio_pal() — all palette types
    (categorical, small-group, scientific, sequential, diverging) are
    accessible from a single function.
  - Added 8 sequential palettes (blue, teal, gray, orange, purple, red,
    green, amber) and 3 diverging palettes (blue_orange, blue_red,
    teal_orange).
  - Added curated small-group variants: duo_warm, duo_cool, trio_bold,
    trio_cool, quad_earth, quad_vivid.
  - Added scientific palettes: okabe_ito, viridis, inferno, plasma.
  - list_ekio_palettes() now validates the type argument and errors on
    invalid types.
  - ekio_orange now has 10 shades (50-900), matching the other color
    scales.

Themes

  - theme_ekio() rebuilt using ggplot2's modular theme_sub_*() helpers
    (requires ggplot2 >= 3.5.0).
  - Added theme_ekio_map() for choropleth and spatial visualizations.

Scales

  - Added continuous scale functions: scale_color_ekio_c() and
    scale_fill_ekio_c() for sequential and diverging palettes.
  - British spelling aliases (scale_colour_ekio_c(),
    scale_colour_ekio_d()) included.

Recipe functions

  - Added high-level chart builders: ekio_histogram(), ekio_lineplot(),
    ekio_scatterplot(), ekio_barplot(), ekio_areaplot().
  - Smart aesthetic detection automatically distinguishes static colors
    from variable mappings and applies appropriate scales.

Tables

  - Added gt_theme_ekio() for professional gt table styling with EKIO
    branding.
  - Fixed gt_theme_ekio() crashing on tables without summary rows or row
    groups.

Datasets

  - Added fuels dataset (global fuel consumption time series).
  - Added brazil_agriculture, brazil_agriculture_states, brazil_gdp, and
    brazil_population datasets.
  - Added ips_brasil dataset (Social Progress Index for Brazilian
    municipalities).

Other

  - Removed dplyr dependency.
  - Removed scales from Imports (not used directly).
  - Removed dead code and obsolete prototype files.
  - Added Shiny palette explorer app (inst/shiny-app/).
  - Added test coverage for recipe functions and gt theme.