Module:tkr-translit

From Linguifex
Jump to navigation Jump to search

This module will transliterate Tsakhur language text. The module should preferably not be called directly from templates or other modules. To use it from a template, use {{xlit}}. Within a module, use Module:languages#Language:transliterate.

For testcases, see Module:tkr-translit/testcases.

Functions

tr(text, lang, sc)
Transliterates a given piece of text written in the script specified by the code sc, and language specified by the code lang.
When the transliteration fails, returns nil.

local gsub = mw.ustring.gsub
local u = require("Module:string utilities").char

local export = {}

function export.tr(text, lang, sc)

	if sc ~= "Cyrl" then
		return nil
	end

	-- Convert capital to lowercase palochka. Lowercase is found in tables.
	text = gsub(text, u(0x4C0), u(0x4CF))

	-- Digraphs with soft and hard sign.
	text = gsub(text, "([хХ])х([ьъ])", "%1%2:")
	text = gsub(text, "[гкхГКХ][ьъ]", {
		["гь"]="h",  ["Гь"]="H",
		["гъ"]="ʁ",  ["Гъ"]="ʁ",
		["кь"]="qӏ", ["Кь"]="Qӏ",
		["къ"]="q:", ["Къ"]="Q:",
		["хь"]="x",  ["Хь"]="X",
		["хъ"]="q",  ["Хъ"]="Q",
	})

	-- Дж and гӏ digraphs.
	text = gsub(gsub(text, "дж", "ǯ"), "Дж", "Ǯ")
	text = gsub(gsub(text, "гӏ", "ɣ"), "Гӏ", "Ɣ")

	-- Geminate consonants.
	text = gsub(text, "[птцчксхПТЦЧКСХ][птцчксх]", {
		["пп"]="п:", ["Пп"]="П:",
		["тт"]="т:", ["Тт"]="Т:",
		["цц"]="ц:", ["Цц"]="Ц:",
		["чч"]="ч:", ["Чч"]="Ч:",
		["кк"]="к:", ["Кк"]="К:",
		["сс"]="с:", ["Сс"]="С:",
		["хх"]="х:", ["Хх"]="Х:",
	})

	-- General consonants.
	text = gsub(text, "[птцчксшщхнлдзжгъбфмрвПТЦЧКСШЩХНЛДЗЖГЪБФМРВ]", {
		["п"]="p",  ["П"]="P",
		["т"]="t",  ["Т"]="T",
		["ц"]="c",  ["Ц"]="C",
		["ч"]="č",  ["Ч"]="Č",
		["к"]="k",  ["К"]="K",
		["с"]="s",  ["С"]="S",
		["ш"]="š",  ["Ш"]="Š",
		["щ"]="š:", ["Щ"]="Š:",
		["х"]="ꭓ",  ["Х"]="Ꭓ",
		["н"]="n",  ["Н"]="N",
		["л"]="l",  ["Л"]="L",
		["д"]="d",  ["Д"]="D",
		["з"]="z",  ["З"]="Z",
		["ж"]="ž",  ["Ж"]="Ž",
		["г"]="g",  ["Г"]="G",
		["ъ"]="ʔ",  ["Ъ"]="ʔ",
		["б"]="b",  ["Б"]="B",
		["ф"]="f",  ["Ф"]="F",
		["м"]="m",  ["М"]="M",
		["р"]="r",  ["Р"]="R",
		["в"]="v",  ["В"]="V",
	})

	-- Palatised consonants.
	text = gsub(text, "([tckdgszxnlTCKDGSZXNL]ӏ?)(:?)([иеёюя])", function (con, gem, vow)
		return con .. "ʲ" .. gem .. (({ ["ё"]="о", ["ю"]="у", ["я"]="a" })[vow] or vow)
	end)

	-- Rounded consonants.
	text = gsub(text, "([tcčkqǯgzšžxꭓʁTCČKQǮGZŠŽXꞳ]ӏ?:?)v", "%1ʷ")

	-- Iotated е.
	text = gsub(text, "%f[%a]е", "je")
	text = gsub(text, "%f[%a]Е", "Je")

	-- Ejective consonants.
	text = gsub(text, "[ptcčkgqPTCČKGQ]ӏ", {
		["pӏ"]="ṗ", ["Pӏ"]="Ṗ",
		["tӏ"]="ṭ", ["Tӏ"]="Ṭ",
		["cӏ"]="c̣", ["Cӏ"]="C̣",
		["čӏ"]="č̣", ["Čӏ"]="Č̣",
		["kӏ"]="ḳ", ["Kӏ"]="Ḳ",
		["qӏ"]="q̇", ["Qӏ"]="Q̇",
	})

	-- Umlaut vowels.
	text = gsub(text, "[аоуАОУ]ь", {
		["аь"]="ä", ["Аь"]="Ä",
		["оь"]="ö", ["Оь"]="Ö",
		["уь"]="ü", ["Уь"]="Ü",
	})

	-- Glottalised vowels.
	text = gsub(text, "([аоуыАОУЫ])ӏ", "%1ˤ")

	-- Long и.
	text = gsub(gsub(text, "ий", "ī"), "Ий", "Ī")

	-- General vowels.
	text = gsub(text, "[йиеэаыоуюяё’ЙИЕЭАЫОУЮЯЁ]", {
		["й"]="j", ["Й"]="J",
		["и"]="i", ["И"]="I",
		["е"]="e", ["Е"]="E",
		["э"]="e", ["Э"]="E",
		["а"]="a", ["А"]="A",
		["ы"]="ɨ", ["Ы"]="Ɨ",
		["о"]="o", ["О"]="O",
		["у"]="u", ["У"]="U",
		["ю"]="ju", ["Ю"]="Ju",
		["я"]="ja", ["Я"]="Ja",
		["ё"]="jo", ["Ё"]="Jo",
		["’"]="ʲ",
	})

	return text

end

return export