l10n_countries 2.1.1

Provides country names translations (for 194 different locales).

2.1.1

FIX

• Reviewed the Slovak (sk) locale against the official list of country names published by the Slovak geodesy and cartography authority, fixing 75 entries:
  • Replaced corrupted characters (e.g. in AZE, DJI, FJI, KNA, GBR) and typos (e.g. in GRC, NLD, CXR, UMI).
  • Fixed factual errors: Morocco was called a principality, French Guiana and the Democratic Republic of the Congo shared their short names with Guyana and the Republic of the Congo, and Eswatini still had its former name.
  • Updated outdated official names (e.g. BOL, BRN, COM, LBY, IRL, BIH) and aligned short names (e.g. GBR, USA, KOR, PRK) and Slovak exonyms (e.g. BES, CUW, MTQ, SLB, VGB) with the official list.
• Fixed short names shared by two different countries in other locales:
  • Serbian (sr): GIN was translated as Guyana (both short and official name).
  • Polish (pl): SSD was translated as Sudan.
  • Albanian (sq): NGA repeated the name of Niger.
  • Hungarian (hu): ASM repeated the name of Samoa.
  • Finnish (fi): VGB and VIR both lacked their British and United States qualifiers.
  • Sundanese (su): TWN repeated the name of China.
• Disambiguated the French (MAF) and Dutch (SXM) parts of Saint Martin, which shared a short or an official name in the Czech (cs), Persian (fa), Croatian (hr), Portuguese (pt), Slovak (sk) and Serbian (sr) locales.
• Fixed other data errors:
  • Croatian (hr): added the missing space in the HKG and MAC official names, and the NLD official name now matches its short name.
  • Portuguese (pt): capitalized the MAF official name.
  • Nepali (ne): removed the unbalanced opening parenthesis from MAC.
  • Dutch (nl): added the missing space in the HMD short name.
  • Trimmed stray whitespace in Japanese (ja), Korean (ko), Polish (pl) and Urdu (ur) names.

2.1.0

FIX

• Renamed NRU in the English locale to "Naoero" ("Republic of Naoero" for the official name), following the United Nations' update of the country's name. Other locales keep their own exonyms (German still reads "Nauru").
• Added the missing NRU self-translation to the Nauruan (na) locale, which previously had no entry for its own country.

IMPROVEMENTS

• availableLocales is now built on first read instead of in the constructor, making a localize() call (mapper construction included) around 15% faster. Mappers are single-use, so every call previously paid for materializing the full 193-locale set even when it was never read.

DOCUMENTATION

• Added a bundled agent skill (skills/l10n-countries-localization), compliant with the Agent Skills specification, installable via dart run skills@ get.
• Replaced the README's inline LLM agent instructions with a pointer to that skill.

2.0.3

TEST

• Integrated dartdoc_test validation for code samples in documentation comments.

DOCUMENTATION

• Fixed the CountriesLocaleMapper.localize example code snippet to instantiate the mapper.

2.0.2

DOCUMENTATION

Docs only, no code changes:

• Added FOSSA status badge to the README.

2.0.1

DOCUMENTATION

Docs only, no code changes:

• Improved CHANGELOG for clarity.
• Added LLM agent instructions to the README.

2.0.0

🎉 Third anniversary and new major release!

NEW FEATURES

• L10N values are now provided in sentence case!
• Tree-shakable via dart define flags.
• Optimized memory usage with lazy locale initialization - translations are now loaded on-demand instead of all at once.
• Reduced initial memory footprint by ~90% - only requested locales are instantiated.
• Added availableLocales getter to query all supported locales without materializing them.
• Single-use design: Mapper instances automatically free memory after localize() is called and cannot be reused. This ensures optimal memory efficiency - create a new instance if you need to localize again.

BREAKING CHANGES

• Single-use design: Mapper instances automatically free memory after localize() is called and cannot be reused. This ensures optimal memory efficiency - create a new instance if you need to localize again.

Previously, localized strings were provided in mixed lowercase (e.g., "islas Malvinas", in Spanish for FLK code) and sentence case. They are now unified and provided in sentence case only (e.g., "Islas Malvinas", in Spanish for FLK code) to preserve capitalization context for proper nouns and ensure immediate compatibility with independent UI labels.

Justification: Capitalization is context-sensitive and cannot be reliably reconstructed from lowercase source strings. By providing values in sentence case, we ensure high-fidelity data for headers and labels. This was also part of discussion in the past.

Migration:

• If you use these values as standalone labels, no action is required (you can also remove your formatter callback, if not needed).
• If you require mid-sentence (inline) text, use the formatter callback to strictly adapt the casing, rather than relying on direct string manipulation.

final localized = mapper.localize(
  isoCodes,
  mainLocale: locale,
  formatter: (_, l10n) => l10n.toLowerCase(), // <-- e.g., inline usage
);

CHORE

• The Dart SDK was bumped to v3.10.4.
• Updated Indonesian country translations.

1.3.1

CHORE

• Lower Dart SDK constraint from ^3.9.2 to ^3.9.0
• Enable and fix 10 new Dart Code Metrics rules (from the 1.32.0: October Update)

1.3.0

FIX

• Corrected Slovak country name for CIV to "Pobrežie Slonoviny".

DOCUMENTATION

• Improved documentation in README.

CHORE

• The Dart SDK was bumped to v3.9.2.

1.2.0

FIX

• Corrected Welsh official country name for Curaçao.
• Corrected Korean localization for British Indian Ocean Territory, Dominica, Mongolia, Wallis and Futuna and South Georgia.

1.1.1

CHORE

• The Dart SDK was bumped to v3.8.1.
• Update German and English translations.

1.1.0

NEW FEATURES

• Add formatter callback for custom translation logic.
• Add official country translations for the Indonesian language.

CHORE

• The Dart SDK was bumped to v3.8.0.
• Code has been formatted with the new Dart formatter.

DOCUMENTATION

• Improved documentation in README.

1.0.0

🎉 First stable release!

DOCUMENTATION

• Improved documentation and example.
• Removed duplicated localization.
• Fixed typos in the README.

0.1.0

• Initial release.