Modül:eve-alfabeçeviri

local export = {}

local u = mw.ustring.char

local MACRON    = u(0x0304)
local DOTABOVE  = u(0x0307)

local tab = { 
	='A', ='a', ='B', ='ʙ', ='W', ='w',
	='E', ='e', ='Jo', ='jo', ='G', ='g',
	='D', ='d', ='I', ='i',  ='Ī', ='ī',
	='J', ='j', ='K', ='k', ='L', ='l',
	='M', ='m', ='N', ='n', ='Ŋ', ='ŋ',
	='O', ='o', ='Ö', ='ö', ='Ö', ='ö',
	='P', ='p', ='R', ='r',	='S', ='s',
	='T', ='t', ='U', ='u', ='Ū', ='ū',
	='F', ='f', ='H', ='h', ='C', ='c',
	='I', ='i', ='Ə', ='ə', ='Ju', ='ju',
	='Ẹ', ='ẹ', -- Not present in the original latinisation
	-- non-native letters
	 ='Z', ='z',	='Z', ='z',
	 ='C', ='c',  ='S', ='s', ='S', ='s', 
	 ='ʺ', ='ʺ', ="’", ="’",
	 
	 -- non-standard letters
	 ='Γ', ='γ', ='Ŋ', ='ŋ', ='Ü', ='ü',
	 ='Ḥ', ='ḥ', ='Q', ='q'							
}


local iotatedTranslit = {
	 = "je",
	 = "ja",
	 = "ji",
	 = "jī",
	 = "Je",
	 = "Ja",
}

local replacements = {
	{ "Ё", 'Jo' },
	{ "ё", 'jo' },
	{ "Ӫ", 'Jö' },
	{ "ӫ", 'jö' },
	{ "Ю", 'Ju' },
	{ "ю", 'ju' },
	
	-- Unfortunately the Cyrillic alphabet doesn't distinguish between ʒe and ʒə
	{ "Де", 'Ʒe' },
	{ "де", 'ʒe' },
	{ "Не", 'Ņe' },
	{ "не", 'ņe' },
	
	{ "Ди", 'Ʒi' },
	{ "ди", 'ʒi' },
	{ "Ни", 'Ņi' },
	{ "ни", 'ņi' },
	
	{ "Дя", 'Ʒa' },
	{ "дя", 'ʒa' },
	{ "Ня", 'Ņa' },
	{ "ня", 'ņa' },
	
	{ "Дj", 'Ʒ' },
	{ "дj", 'ʒ' },
	{ "Нj", 'Ņ' },
	{ "нj", 'ņ' },
	
	-- The following is non-standard but supposedly used in non-standard spelling and dialect forms
	{ "Дь", 'Ʒ' },
	{ "дь", 'ʒ' },
	{ "Нь", 'Ņ' },
	{ "нь", 'ņ' },
	
	{ "Нг", 'Ŋ' },
	{ "нг", 'ŋ' },
}
 
function export.tr(text, lang, sc)
	local ugsub, str_gsub = mw.ustring.gsub, string.gsub
	local UTF8char = '*'
	
	for i, replacement in ipairs(replacements) do
		text = str_gsub(text, unpack(replacement))
	end
	
	-- е after a vowel or at the beginning of a word becomes ye
	-- Again, the Cyrillic alphabet doesn't distinguish between je and jə
	text = ugsub(text,
		"(?)()", 
		function(preceding, iotated)
			return preceding .. iotatedTranslit
		end)
	
	text = ugsub(text, "^", iotatedTranslit)
	
	text = ugsub(text, "()()",
		function(preceding, iotated)
			return preceding .. iotatedTranslit
		end)
	
	text = str_gsub(text, UTF8char, tab)
	
	return text
end

return export