local export = {}
local yodated_v = "ЯяЮюЁё"
local vowels = "үҮұҰеЕэЭәӘөӨАаоОЫыІі" .. "اىوۇە" .. yodated_v
local consonants = "БбВвГгҒғДдЖжЗзЙйКкҚқЛлМмНнҢңПпРрСсТтФфХхҺһЦцЧчШшЩщЪъЬь" .. "بپتجچحدرزسشعفقكگڭلمنۆھ"
local semiv = "УуИи" .. "يۋ" .. "IiyUuw" -- these letters act irregularly so deal with them separately
local U = mw.ustring.char
local gsub = mw.ustring.gsub
local hamzah = U(0x674) -- Kazakh High Hamza
local sukuun = U(0x652) -- sukuun is not used in Kazakh arabic orthography, its only used here to supress vowel fronting when respelling terms in quotation templates
local tt = {}
tt["Cyrl"] = {
["ү"]="ü",["Ү"]="Ü",["т"]="t",["Т"]="T",["һ"]="h",["Һ"]="H",["р"]="r",["Р"]="R"
,["ф"]="f",["Ф"]="F",["ш"]="š",["Ш"]="š",["н"]="n",["Н"]="N",["п"]="p",["П"]="P"
,["й"]="y",["Й"]="Y",["л"]="l",["Л"]="L",["з"]="z",["З"]="Z",["е"]="e",["Е"]="E"
,["г"]="g",["Г"]="G",["б"]="b",["Б"]="B",["у"]="uw",["У"]="Uw",["с"]="s",["С"]="S"
,["х"]="x",["Х"]="X",["ы"]="ı",["Ы"]="I",["м"]="m",["М"]="M",["о"]="o",["О"]="O"
,["и"]="ıy",["И"]="Iy",["ж"]="j",["Ж"]="J",["к"]="k",["К"]="K",["д"]="d",["Д"]="D"
,["в"]="v",["В"]="V",["а"]="a",["А"]="A",["ң"]="ŋ",["Ң"]="Ŋ",["ә"]="ä",["Ә"]="Ä"
,["ұ"]="u",["Ұ"]="U",["қ"]="q",["Қ"]="Q",["ғ"]="ğ",["Ғ"]="Ğ"
,["і"]="i",["І"]="İ",["ө"]="ö",["Ө"]="Ö"
--mostly in loanwords from Russian
,["ё"]="yo",["Ё"]="Yo",["ц"]="c",["Ц"]="C",["ч"]="č",["Ч"]="Č",["щ"]="šč",["Щ"]="Šč"
,["ъ"]="ʺ",["Ъ"]="ʺ",["ь"]="ʹ",["Ь"]="ʹ",["э"]="è",["Э"]="È",["ю"]="yuw",["Ю"]="Yuw"
,["я"]="ya",["Я"]="Ya"};
tt["Arab"] = {
-- consonants
["م"] = "m", ["ن"] = "n", ["د"] = "d", ["ت"] = "t",
["ب"] = "b", ["پ"] = "p", ["ف"] = "f", ["ق"] = "q",
["ك"] = "k", ["ڭ"] = "ŋ", ["گ"] = "g", ["ع"] = "ğ",
["ھ"] = "h", ["ح"] = "x", ["چ"] = "č",
["ج"] = "j", ["ز"] = "z", ["س"] = "s",
["ش"] = "š", ["ر"] = "r", ["ل"] = "l",
["ي"] = "ıy", ["ۋ"] = "uw", ["ۆ"] = "v",
-- vowels
["ا"] = "a", ["ە"] = "e", ["ى"] = "ı",
["و"] = "o", ["ۇ"] = "u", [hamzah] = "", [sukuun] = "",
-- punctuation
["؟"]="?",
["،"]=",",
["؛"]=";",
["ـ"]="-"
}
local function vowelfronting(text)
text = gsub(text, "ا", "ä")
text = gsub(text, "ە", "e")
text = gsub(text, "ى", "i")
text = gsub(text, "ي", "iy")
text = gsub(text, "و", "ö")
text = gsub(text, "ۇ", "ü")
text = gsub(text, "ۋ", "üw")
text = gsub(text, "Я", "Йә")
text = gsub(text, "я", "йә")
text = gsub(text, "Ё", "Йө")
text = gsub(text, "ё", "йө")
return text
end
function export.tr(text, lang, sc)
if type(text) == "table" then
options = {}
text, script = text.args[1], text.args[2]
end
if (sc == "Cyrl") then
text = gsub(text, '([юЮ])', function(c)
local yodated = {
["ю"]="йу",["Ю"]="Йу"}
return yodated[c]
end)
text = gsub(text,
"([АОҰЭЯЁЮЕЪЬаоұэяёюеьъ][́̀]?)([ЕеЭэ])",
function(a, e)
local iotated = {
['е'] = 'йе', ['Е'] = 'ЙЕ',
['э'] = 'е', ['Э'] = 'Е', -- yodate it now bc removing it later fixes issues
}
return a .. iotated[e]
end)
-- add vowel
local frontv = {
['И'] = 'İy',
['и'] = 'iy',
['У'] = 'Üw',
['у'] = 'üw',
}
local backv = {
['И'] = 'Iy',
['и'] = 'ıy',
['У'] = 'Uw',
['у'] = 'uw',
}
local defaultv = {
['И'] = 'İy',
['и'] = 'iy',
['У'] = 'Uw',
['у'] = 'uw',
}
text = gsub(text, '([^' .. consonants .. semiv .. '])([Уу])',
function(v, s)
local semiv = {
['У'] = 'W',
['у'] = 'w',
}
return v .. semiv[s]
end)
text = gsub(text, '^([Уу])([^' .. consonants .. '])',
function(s, v)
local semiv = {
['У'] = 'W',
['у'] = 'w',
}
return semiv[s] .. v
end)
-- match the fronting of the following vowel, if there is one
text = gsub(text, '([ИиУу])([' .. vowels ..'])',
function(s, v)
if mw.ustring.match(v, '[үҮеЕэЭәӘөӨІі]') then
return frontv[s] .. v
else
return backv[s] .. v
end
end)
-- otherwise, match the fronting of the previous vowel
text = gsub(text, '([' .. vowels .. "КкГгҚқҒғ][" .. consonants .. semiv .. ']*)([ИиУу])',
function(v, s)
if mw.ustring.match(v, '[үҮеЕэЭәӘөӨІіГгКк]') then
return v .. frontv[s]
else if mw.ustring.match(v, '[ұҰАаоОЫыҒғҚқ]') then
return v .. backv[s]
else
return v .. defaultv[s]
end
end
end)
--[=[ EXCEPTION Каир = Kair
this only ever happens in Russian loanwords so we wont bother with vowel harmony ]=]
text = gsub(text, '([' .. vowels .. '])([Ии])([' .. consonants .. '])',
function(v, s, c)
local semiv = {
['И'] = 'İ',
['и'] = 'i',
}
return v .. semiv[s] .. c
end)
text = gsub(text, "([İiIı])[Yy]([ЯяЮюЁёЭэ])", "%1%2")
text = gsub(text, "[Йй]([эЭ])", "%1")
if mw.ustring.match(text, "[" .. vowels .. "ГгКк][" .. consonants .. semiv .. "]*[ЯяЮюЁё]") then
text = gsub(text, "(" .. "[үҮеЕэЭәӘөӨІіГгКк]" .. "[" .. consonants .. semiv .. "ÎîÛû]*[ЯяЮюЁё])", vowelfronting) else
text = gsub(text, "(" .. "[ЯяЮюЁё]" .. "[" .. consonants .. semiv .. "ÎîÛû]*" .. "[үҮеЕэЭәӘөӨІіГгКк])", vowelfronting)
end
end
if (sc == "Arab") then
-- mark fronting
text = gsub(text, U(0x621), hamzah)
--[=[ Many government sources in China and Kazakhstan use regular hamza
so we'll allow it here; But ideally, it should be replaced in entries.
]=]
text = gsub(text, "([" .. vowels .. consonants .. semiv .. "?]+)([ەكگ])", hamzah .. "%1%2")
text = gsub(text, "([^" .. vowels .. consonants .. semiv .. "]+)([ەكگ])", "%1" .. hamzah .. "%2")
text = gsub(text, "^([ەكگ])", hamzah .. "%1")
-- change ۋي in open syllables
text = gsub(text, "([^" .. consonants .. semiv .. "])([ۋي])([" .. vowels .. "])",
function(c, s, v)
local addv = {
['ۋ'] = 'w',
['ي'] = 'y',
}
return c .. addv[s] .. v
end)
text = gsub(text, "([" .. vowels .. "])([ۋي])",
function(v, s)
local addv = {
['ۋ'] = 'w',
['ي'] = 'y',
}
return v .. addv[s]
end)
text = gsub(text, "^([ۋي])([^" .. consonants .. semiv .. "])",
function(s, c)
local addv = {
['ۋ'] = 'w',
['ي'] = 'y',
}
return addv[s] .. c
end)
--[=[text = gsub(text, "([" .. consonants .. semiv .. "])([ۋي])([" .. consonants .. semiv .. "])",
function(c, s, v)
local addv = {
['ۋ'] = 'û',
['ي'] = 'î',
}
return c .. addv[s] .. v
end)
text = gsub(text, "([^" .. vowels .. consonants .. semiv .. "])([ۋي])([" .. consonants .. semiv .. "])",
function(c, s, v)
local addv = {
['ۋ'] = 'û',
['ي'] = 'î',
}
return c .. addv[s] .. v
end)
text = gsub(text, "^([ۋي])([" .. consonants .. "])",
function(s, v)
local addv = {
['ۋ'] = 'û',
['ي'] = 'î',
}
return addv[s] .. v
end)]=]
-- vowel harmony
text = gsub(text, "(" .. hamzah .. "[" .. consonants .. vowels .. semiv .. "îû]+)", vowelfronting)
-- ق and ع break vowel harmony
text = gsub(text,
"([قع])([äiöü])",
function(c, v)
local defronted = {
['ä'] = 'a', ['i'] = 'ı',
['ö'] = 'o', ['ü'] = 'u',
}
return c .. defronted[v]
end)
text = gsub(text,
"([äiöü])([قع])",
function(v, c)
local defronted = {
['ä'] = 'a', ['i'] = 'ı',
['ö'] = 'o', ['ü'] = 'u',
}
return defronted[v] .. c
end)
end
return mw.ustring.toNFC(mw.ustring.gsub(text, '.', tt[sc]))
end
return export
