Module:mn-IPA

local export = {}
local mn = require("Module:mn-common")
local gsub = mw.ustring.gsub
local u = require("Module:string/char")

local vowels1 = {
	 = 'a',  = 'ɔ',  = 'ʊ',
	 = 'e',  = 'ɵ',  = 'u',
	 = 'i',  = 'i',  = 'i', --й very occasionally used at the start of loanwords
	 = 'ja',  = 'jʊ',  = 'jɔ',  = 'jɵ',
	 = '',  = 'ə',
}

local vowels2 = {
	 = 'aː',  = 'ɛː',  = 'jaː',  = 'jɛː',  = 'ja',  = 'ɒː',  = 'aʊ̯',  = 'jaʊ̯',
	 = 'oː',  = 'ɔi̯',  = 'jɔː',  = 'jɔi̯',  = 'jɔ',  = 'ʊ̯a',  = 'joʊ',
	 = 'ʊː',  = 'ʊi̯',  = 'u̯a',  = 'jʊː',  = 'jʊi̯',  = 'u̯iː',
	 = 'eː',  = 'ei̯',  = 'jeː',  = 'jei̯',  = 'ia',  = 'eɔ',  = 'iʊ̯',  = 'iuː',
	 = 'ɵː',  = 'ɵi̯',  = 'jɵ',
	 = 'uː',  = 'ui̯',  = 'juː',  = 'jui̯',
	 = 'iː',  = 'iː',  = 'i̯aː',  = 'i̯oː',  = 'i̯ʊː',  = 'i̯eː',  = 'i̯ɵː',  = 'i̯uː',
}

local vowels3 = {
	 = 'jaː',  = 'jɔː',  = 'jeː',
	 = 'jɛː',  = 'u̯ɛː',  = 'jiː',
	 = 'aʊ̯a',
}

local consonants1 = {
	 = 'pʰ',  = 'p',  = 'tʰ',  = 't',  = 'kʰ',  = 'k',
	 = 't͡sʰ',  = 't͡s',  = 't͡ɕʰ',  = 't͡ɕ',
	 = 'f',  = 'w̜',  = 's',  = 'ɕ',  = 'ɕ',  = 'x',
	 = 'm',  = 'n',
	 = 'r',  = 'ɮ',
	 = 'ʲ',
}

local consonants2 = {
	 = 'b',
	 = 'b',
	 = 't͉.t',
	 = 'ɮː',  = 'ɬʰ',
	 = 'mː',
	 = 'mp',  = 'nː',
	 = 'r̥kʰ',  = 'r̥pʰ',  = 'r̥tʰ',  = 'r̥t͡sʰ',  = 'r̥t͡ɕʰ',
	 = 'sː',
}

local double_consonants = {
	 = 'п',  = 'т',  = 'к',  = 'г',
	 = 'ц',  = 'з',  = 'ч',  = 'ж',
	 = 'ф',  = 'в',  = 'ш',  = 'щ',  = 'х',
	 = 'р',
}

local palatalisation = {
	 = 'пʲө',  = 'пʲу',  = 'пʲо',  = 'пʲа',
	 = 'бʲө',  = 'бʲу',  = 'бʲо',  = 'бʲа',
	 = 'тʲө',  = 'тʲу',  = 'тʲо',  = 'тʲа',
	 = 'дʲө',  = 'дʲу',  = 'дʲо',  = 'дʲа',
	 = 'кʲө',  = 'кʲу',  = 'кʲо',  = 'кʲа',
	 = 'гʲө',  = 'гʲу',  = 'гʲо',  = 'гʲа',
	 = 'фʲө',  = 'фʲу',  = 'фʲо',  = 'фʲа',
	 = 'вʲө',  = 'вʲу',  = 'вʲо',  = 'вʲа',
	 = 'хʲө',  = 'хʲу',  = 'хʲо',  = 'хʲа',
	 = 'мʲө',  = 'мʲу',  = 'мʲо',  = 'мʲа',
	 = 'нʲө',  = 'нʲу',  = 'нʲо',  = 'нʲа',
	 = 'рʲө',  = 'рʲу',  = 'рʲо',  = 'рʲа',
	 = 'лʲө',  = 'лʲу',  = 'лʲо',  = 'лʲа',
}


function export.to_IPA(text)

	local orig_text = text
	text = mw.ustring.lower(text)
	
	text = gsub(text, '-', '')
	
	text = gsub(text, "ы$", "ы~")
	text = gsub(text, "ы(%s)", "ы~%1")
	
	--primary stress
	-- U+0301 COMBINING ACUTE ACCENT, U+02C8 MODIFIER LETTER VERTICAL LINE
	--secondary stress
	-- U+0300 COMBINING GRAVE ACCENT, U+02C8 MODIFIER LETTER LOW VERTICAL LINE
	text = gsub(text, "'", u(0x02C8)) -- can use apostrophe for convenience
	
	if not mw.ustring.match(text, u(0x02C8)) then
		if not mw.ustring.match(text, '') then
			text = u(0x02C8) .. text
		end
		text = gsub(text, '()()'..u(0x0301), '%1'..u(0x02C8)..'%2')
		text = gsub(text, '(%*?)()'..u(0x0301), u(0x02C8)..'%1%2')
		text = gsub(text, '()()'..u(0x0300), '%1'..u(0x02CC)..'%2')
		text = gsub(text, '(%*?)()'..u(0x0300), u(0x02CC)..'%1%2')
	end
	for k, v in pairs(double_consonants) do
		text = gsub(text, k, v)
		text = gsub(text, k, v) --remove double consonants
	end
	for k, v in pairs(palatalisation) do
		text = gsub(text, k, v)
		text = gsub(text, k, v) --convert iotated vowels following consonants
	end
	for k, v in pairs(vowels3) do
		text = gsub(text, k, v)
		text = gsub(text, k, v)
		text = gsub(text, k, v)
	end
	for k, v in pairs(vowels2) do
		text = gsub(text, k, v)
		text = gsub(text, k, v)
	end
	for k, v in pairs(vowels1) do
		text = gsub(text, k, v)
	end
	
	for k, v in pairs(consonants2) do
		text = gsub(text, k, v)
		text = gsub(text, k, v)
	end
	
	text = gsub(text, '%*?', consonants1)
	text = gsub(text, '()t͡ɕ()', '%1d͡ʑ%2')
	text = gsub(text, '()p()', '%1β%2')
	text = gsub(text, '()k()', '%1ɣ%2')
	text = gsub(text, 'n(*)%f', 'm%1')
	text = gsub(text, 'n%f', 'ŋ')
	text = gsub(text, 'r%f', 'r̥')

	text = gsub(text, '()i%f', '%1ⁱ')
	
	if #mn.syllables( orig_text ) ~= 1 then
		text = gsub(text, '%f', '')
	end
	
	text = gsub(text, 'ⁱ', 'i')
	text = gsub(text, 'ʲj', 'j')

	--put ~ at the end of respelling to prevent vowel cutoff
	text = gsub(text, '~', '')

	return text

end

-- only works with Cyrillic Mongolian
function export.show(frame)
	local args = frame:getParent().args
	local page_title = mw.title.getCurrentTitle().text
	local text = args or page_title
	local qualifier = args or nil
	local lang = require("Module:languages").getByCode("mn")
	local script = require("Module:scripts").getByCode("Cyrl")
	
	local syllables = mn.syllables(text)
	local transcription = export.to_IPA(text)
	local IPA = "<li>" .. (qualifier and require("Module:qualifier").format_qualifier{qualifier} .. " " or "") ..
		require("Module:IPA").format_IPA_full {
		lang = lang,
		items = {{ pron = "/" .. transcription .. "/" }},
	} .. "</li>"
	local hyphenation = "<li>" .. require("Module:hyphenation").format_hyphenations { 
		lang = lang,
		hyphs = { { hyph = syllables } },
		sc = script,
		caption = "]",
		} .. mw.ustring.char( 0x2002 ) .. "(" .. #syllables .. " syllable"
	if #syllables >= 2 then hyphenation = hyphenation .. "s" end
	hyphenation = hyphenation .. ")</li>"

	return "<ul>" .. IPA .. hyphenation .. "</ul>"
end

return export