Changes in version 0.3.0                        

New Features

  - Bundled party data: Package now includes pre-scraped color and logo
    data from the Party Facts Wikipedia dataset, providing instant
    lookups without Wikipedia scraping. This dramatically improves
    performance and reduces load on Wikipedia servers.

  - New dataset party_data - bundled party colors and logos from
    Wikipedia with columns: url, color, all_colors, logo_url, and
    last_updated. Includes all parties from the Party Facts Wikipedia
    dataset with available color data.

  - New parameter use_cache (default TRUE) in get_party_color(),
    get_party_logo(), and get_party_info() to prefer bundled data over
    live scraping. Set to FALSE to always scrape fresh data from
    Wikipedia.

  - New script data-raw/update.R to regenerate bundled data from
    Wikipedia. Package maintainers can use this to periodically update
    the cached data.

Improvements

  - Faster performance: Bundled data lookups are ~100x faster than live
    scraping for parties in the cache.

  - Reduced Wikipedia server load: Default behavior now uses cached
    data, with graceful fallback to live scraping for parties not in the
    bundled dataset.

  - Graceful fallback: If a party is not in bundled data, functions
    automatically fall back to live scraping, ensuring full backward
    compatibility.

  - Offline capability: Cached parties can be accessed without internet
    connection.

Documentation

  - Added comprehensive documentation for party_data dataset.

  - Updated all function examples to demonstrate both cached and live
    scraping modes.

  - Added new test suite test-bundled_data.R with 20+ tests for cache
    functionality.

  - Updated vignette with section on bundled data vs. live scraping.

Internal

  - Added LazyData: true to DESCRIPTION for efficient data loading.

  - Added cache lookup helper functions: lookup_cached_color(),
    lookup_cached_logo(), lookup_cached_info().

  - Updated all integration tests to explicitly use use_cache = FALSE
    for testing live scraping functionality.

                 Changes in version 0.2.0 (2026-01-27)                  

New Features

  - New function get_party_color() - vectorized color extraction with
    support for multiple colors (all_colors parameter) and color
    normalization.

  - New function get_party_logo() - extract party logo URLs from
    Wikipedia pages.

  - New function get_party_info() - extract both color and logo in a
    single call for better efficiency.

  - New Partyfacts integration:
    
      - get_partyfacts_wikipedia() - download the Partyfacts Wikipedia
        dataset
      - lookup_party_url() - search for parties by name and country
      - get_party_color_by_name() - convenience function combining
        lookup and color extraction
      - clear_partycolor_cache() - clear cached Partyfacts data

Improvements

  - Functions are now fully vectorized and work seamlessly with
    dplyr::mutate().

  - Better error handling - invalid URLs return NA instead of throwing
    errors.

  - Color normalization converts named colors (e.g., "red") to hex codes
    and expands 3-digit hex codes to 6 digits.

  - Added comprehensive test suite using testthat.

  - Complete vignette with examples and best practices.

  - Updated to use modern tidyverse patterns (replaced gather/spread
    with pivot_longer/pivot_wider).

Breaking Changes

  - The original wikipedia_party_color() function is now deprecated. Use
    get_party_color() instead.

  - The package now requires R >= 4.1.0.

Documentation

  - Added detailed function documentation with examples.

  - New vignette: "Getting Started with partycoloR".

  - Updated README with feature overview and quick start guide.

                        Changes in version 0.1.0                        

  - Initial release with wikipedia_party_color() function.

  - Extract party colors from English Wikipedia party pages.