Module:ábḫ-ipa

From Linguifex
Revision as of 00:22, 28 February 2020 by Thisimpliesthis (talk | contribs)
Jump to navigation Jump to search


local export = {}

function export.convert(frame)
	local word = type(frame) == "table" and frame.args[1] or frame
	
	-- go from orthography to one-to-one 
	word = string.gsub(word, "mb", "B")
	word = string.gsub(word, "nd", "D")
	word = string.gsub(word, "ng", "G")
	word = string.gsub(word, "mp", "P")
	word = string.gsub(word, "ntt", "Tt")
	word = string.gsub(word, "nkk", "Kk")
	word = string.gsub(word, "jh", "J")
	word = string.gsub(word, "ḫ", "H")
	word = string.gsub(word, "ii", "ī")
	word = string.gsub(word, "íi", "î")
	word = string.gsub(word, "ií", "ǐ")
	word = string.gsub(word, "ei", "ē")
	word = string.gsub(word, "éi", "ê")
	word = string.gsub(word, "eí", "ě")
	word = string.gsub(word, "ou", "ō")
	word = string.gsub(word, "óu", "ô")
	word = string.gsub(word, "oú", "ǒ")
	word = string.gsub(word, "oi", "I")
	word = string.gsub(word, "ói", "Í")
	word = string.gsub(word, "aai", "Ā")
	word = string.gsub(word, "áai", "Â")
	word = string.gsub(word, "aái", "Ǎ")
	word = string.gsub(word, "ai", "A")
	word = string.gsub(word, "ái", "Á")
	word = string.gsub(word, "aao", "Ō")
	word = string.gsub(word, "áao", "Ô")
	word = string.gsub(word, "aáo", "Ǒ")
	word = string.gsub(word, "ao", "O")
	word = string.gsub(word, "áo", "Ó")
	word = string.gsub(word, "aa", "ā")
	word = string.gsub(word, "áa", "â")
	word = string.gsub(word, "aá", "ǎ")
	
	word = mw.text.split(word, "", true)
	
	result = {}
	
	-- put in syllable boundary marks
	for i,val in ipairs(word) do
		if word[i-1] == nil then
			if mw.ustring.match(word[i], "[íîǐéêěóôǒÍáâǎÁÂǍÓÔǑ]") then
				table.insert(result, "ˈ")
				table.insert(result, val)
			elseif mw.ustring.match(word[i], "[mnjBDGPpTtKkbzdsJHhry]") and mw.ustring.match(word[i+1], "[íîǐéêěóôǒÍáâǎÁÂǍÓÔǑ]") then
				table.insert(result, "ˈ")
				table.insert(result, val)
			else
				table.insert(result, ".")
				table.insert(result, val)
			end
		elseif mw.ustring.match(word[i], "[iīeēoōIaāAĀOŌ]") then
			if mw.ustring.match(word[i-1], "[iīeēoōIaāAĀOŌíîǐéêěóôǒÍáâǎÁÂǍÓÔǑ]") then
				table.insert(result, ".")
				table.insert(result, val)
			else
				table.insert(result, val)
			end
		elseif mw.ustring.match(word[i], "[íîǐéêěóôǒÍáâǎÁÂǍÓÔǑ]") then
			if mw.ustring.match(word[i-1], "[iīeēoōIaāAĀOŌíîǐéêěóôǒÍáâǎÁÂǍÓÔǑ]") then
				table.insert(result, "ˈ")
				table.insert(result, val)
			else
				table.insert(result, val)
			end
		else
			if word[i+1] == nil then
				table.insert(result, val)
			elseif mw.ustring.match(word[i+1], "[iīeēoōIaāAĀOŌ]") then
				table.insert(result, ".")
				table.insert(result, val)
			elseif mw.ustring.match(word[i+1], "[íîǐéêěóôǒÍáâǎÁÂǍÓÔǑ]") then
				table.insert(result, "ˈ")
				table.insert(result, val)
			else
				table.insert(result, val)
			end
		end
	end
	
	-- isolate syllable boundary marks
	stresses = string.gsub(table.concat(result), "[^.ˈ]", "") 
	syllables = string.len(stresses)
	stresses = mw.text.split(stresses, "", true)
	
	-- compute position of stress
	pos = 0
	for i,val in ipairs(word) do
		if mw.ustring.match(word[i], "ˈ") then
			pos = i
		end
	end
	
	newStresses = {}
	
	-- mark secondary stress
	for i,val in ipairs(stresses) do
		if mw.ustring.match(word[i], ".") and i ~= syllables then
			if (i - pos) % 2 == 0 then
				table.insert(newStresses, "ˌ")
			else
				if not (i == 1 and mw.ustring.match(result[1], ".")) then
					table.insert(newStresses, val)
				end
			end
		else
			table.insert(newStresses, val)
		end
	end
	
	-- look through word and replace the correct .'s with ˌ's
	newWord = {}
	index = 1
	for val in words do
		if mw.ustring.match(val, "[.ˈ]") then
			table.insert(newWord, newStresses[index])
			index = index + 1
		else
			table.insert(newWord, val)
		end
	end
	
	word = table.concat(newWord)
	
	-- go from one-to-one phonemic map to IPA
	word = string.gsub(word, "B", "ᵐb")
	word = string.gsub(word, "D", "ⁿd")
	word = string.gsub(word, "j", "ɲ")
	word = string.gsub(word, "G", "ᵑg")
	word = string.gsub(word, "P", "mp")
	word = string.gsub(word, "T", "nt")
	word = string.gsub(word, "K", "nk")
	word = string.gsub(word, "b", "β")
	word = string.gsub(word, "z", "θ")
	word = string.gsub(word, "d", "ð")
	word = string.gsub(word, "s", "ɕ")
	word = string.gsub(word, "J", "ʑ")
	word = string.gsub(word, "H", "x")
	word = string.gsub(word, "h", "ɦ")
	word = string.gsub(word, "r", "ɺ")
	word = string.gsub(word, "y", "j")
	word = string.gsub(word, "ī", "iː")
	word = string.gsub(word, "î", "îː")
	word = string.gsub(word, "ǐ", "ǐː")
	word = string.gsub(word, "ē", "eːɪ̯")
	word = string.gsub(word, "ê", "êːɪ̯")
	word = string.gsub(word, "ě", "ěːɪ̯")
	word = string.gsub(word, "o", "ɔ")
	word = string.gsub(word, "ó", "ɔ́")
	word = string.gsub(word, "ō", "ɔːʊ̯")
	word = string.gsub(word, "ô", "ɔ̂ːʊ̯")
	word = string.gsub(word, "ǒ", "ɔ̌ːʊ̯")
	word = string.gsub(word, "I", "ɔɪ̯")
	word = string.gsub(word, "Í", "ɔ́ɪ̯")
	word = string.gsub(word, "ā", "aː")
	word = string.gsub(word, "â", "âː")
	word = string.gsub(word, "ǎ", "ǎː")
	word = string.gsub(word, "A", "aɪ̯")
	word = string.gsub(word, "Á", "áɪ̯")
	word = string.gsub(word, "Ā", "aːɪ̯")
	word = string.gsub(word, "Â", "âːɪ̯")
	word = string.gsub(word, "Ǎ", "ǎːɪ̯")
	word = string.gsub(word, "O", "aɔ̯")
	word = string.gsub(word, "Ó", "áɔ̯")
	word = string.gsub(word, "Ō", "aːɔ̯")
	word = string.gsub(word, "Ô", "âːɔ̯")
	word = string.gsub(word, "Ǒ", "ǎːɔ̯")
	
	return "/" .. word .. "/"
end

return export