158 lines
3.1 KiB
JavaScript
158 lines
3.1 KiB
JavaScript
/**
|
||
Language-specific replacement rules.
|
||
|
||
Keys are BCP-47 language tags (or their prefixes).
|
||
|
||
Values are arrays of [from, to] replacement pairs.
|
||
|
||
These replacements take precedence over the default replacements when a matching locale is specified.
|
||
*/
|
||
|
||
// Shared Swedish/Finnish replacements
|
||
const swedishFinnishReplacements = [
|
||
['ä', 'a'],
|
||
['Ä', 'A'],
|
||
['ö', 'o'],
|
||
['Ö', 'O'],
|
||
['å', 'a'],
|
||
['Å', 'A'],
|
||
];
|
||
|
||
// Shared Danish/Norwegian replacements
|
||
const danishNorwegianReplacements = [
|
||
['æ', 'ae'],
|
||
['Æ', 'Ae'],
|
||
['ø', 'oe'],
|
||
['Ø', 'Oe'],
|
||
['å', 'aa'],
|
||
['Å', 'Aa'],
|
||
];
|
||
|
||
const localeReplacements = {
|
||
// Swedish
|
||
sv: swedishFinnishReplacements,
|
||
|
||
// Finnish
|
||
fi: swedishFinnishReplacements,
|
||
|
||
// Danish
|
||
da: danishNorwegianReplacements,
|
||
|
||
// Norwegian Bokmål
|
||
nb: danishNorwegianReplacements,
|
||
|
||
// German
|
||
de: [
|
||
['ä', 'ae'],
|
||
['Ä', 'Ae'],
|
||
['ö', 'oe'],
|
||
['Ö', 'Oe'],
|
||
['ü', 'ue'],
|
||
['Ü', 'Ue'],
|
||
['ß', 'ss'],
|
||
['ẞ', 'Ss'],
|
||
],
|
||
|
||
// Turkish
|
||
tr: [
|
||
['â', 'a'],
|
||
['Â', 'A'],
|
||
['ö', 'o'],
|
||
['Ö', 'O'],
|
||
['ü', 'u'],
|
||
['Ü', 'U'],
|
||
],
|
||
|
||
// Hungarian
|
||
hu: [
|
||
['ű', 'u'],
|
||
['Ű', 'U'],
|
||
['ö', 'o'],
|
||
['Ö', 'O'],
|
||
['ü', 'u'],
|
||
['Ü', 'U'],
|
||
['á', 'a'],
|
||
['Á', 'A'],
|
||
['é', 'e'],
|
||
['É', 'E'],
|
||
['í', 'i'],
|
||
['Í', 'I'],
|
||
['ó', 'o'],
|
||
['Ó', 'O'],
|
||
['ú', 'u'],
|
||
['Ú', 'U'],
|
||
],
|
||
|
||
// Bulgarian (Streamlined System)
|
||
bg: [
|
||
['х', 'h'],
|
||
['Х', 'H'],
|
||
['щ', 'sht'],
|
||
['Щ', 'Sht'],
|
||
['ъ', 'a'],
|
||
['Ъ', 'A'],
|
||
['ь', 'y'],
|
||
['Ь', 'Y'],
|
||
// The Russian ъе digraph does not apply in Bulgarian.
|
||
['ъе', 'ae'],
|
||
['Ъе', 'Ae'],
|
||
['ъЕ', 'aE'],
|
||
['ЪЕ', 'AE'],
|
||
],
|
||
|
||
// Serbian
|
||
sr: [
|
||
['х', 'h'],
|
||
['Х', 'H'],
|
||
['ђ', 'dj'],
|
||
['Ђ', 'Dj'],
|
||
['đ', 'dj'],
|
||
['Đ', 'Dj'],
|
||
['џ', 'dz'],
|
||
['Џ', 'Dz'],
|
||
['љ', 'lj'],
|
||
['Љ', 'Lj'],
|
||
['њ', 'nj'],
|
||
['Њ', 'Nj'],
|
||
['ћ', 'c'],
|
||
['Ћ', 'C'],
|
||
['ч', 'ch'],
|
||
['Ч', 'Ch'],
|
||
['ш', 'sh'],
|
||
['Ш', 'Sh'],
|
||
['ж', 'zh'],
|
||
['Ж', 'Zh'],
|
||
],
|
||
};
|
||
|
||
// Convert all locale replacements to Maps at module load time
|
||
for (const locale of Object.keys(localeReplacements)) {
|
||
localeReplacements[locale] = new Map(localeReplacements[locale]);
|
||
}
|
||
|
||
export default localeReplacements;
|
||
|
||
// A combining mark belongs to the letter before it, so marks are skipped when checking the end of a word.
|
||
const wholeWordPattern = word => new RegExp(String.raw`(?<![\p{Letter}\p{Mark}])${word}(?!\p{Mark}*\p{Letter})`, 'gv');
|
||
const wordEndPattern = word => new RegExp(String.raw`${word}(?!\p{Mark}*\p{Letter})`, 'gv');
|
||
|
||
/**
|
||
Language-specific replacements that depend on the surrounding text.
|
||
|
||
Values are arrays of [pattern, replacement] pairs. They are applied after the custom replacements and before the other replacements.
|
||
*/
|
||
export const localeContextReplacements = {
|
||
// Bulgarian
|
||
bg: [
|
||
// The country name is a fixed exception (България → Bulgaria, not Balgaria)
|
||
[wholeWordPattern('България'), 'Bulgaria'],
|
||
[wholeWordPattern('БЪЛГАРИЯ'), 'BULGARIA'],
|
||
[wholeWordPattern('българия'), 'bulgaria'],
|
||
|
||
// Word-final ия is ia (София → Sofia)
|
||
[wordEndPattern('ия'), 'ia'],
|
||
[wordEndPattern('Ия'), 'Ia'],
|
||
[wordEndPattern('ИЯ'), 'IA'],
|
||
],
|
||
};
|