Modul:sa-utilities

Dokumentasi untuk modul ini dapat dibuat di Modul:sa-utilities/doc

local export = {}

export.consonant_list = "kKgGNcCjJYwWqQRtTdDnpPbBmyrlLvSzsh"
export.consonant = "[" .. export.consonant_list .. "]"
export.accent = "[/\\]"
export.vowel_list = "aAiIuUfFxXeEoO"
export.vowel = "[" .. export.vowel_list .. "]"
export.vowel_with_accent = export.vowel .. export.accent .. "?"

local U = mw.ustring.char
local match = mw.ustring.match
local gsub = mw.ustring.gsub
local sub = mw.ustring.sub
local lower = mw.ustring.lower
local upper = mw.ustring.upper

local function ends_with(text, pattern)
	return match(text, pattern .. "$")
end

local function starts_with(text, pattern)
	return match(text, "^" .. pattern)
end

export.up_one_grade = {
	['a'] = 'A', ['A'] = 'A', ['a/'] = 'A/', ['A/'] = 'A/', ['a\\'] = 'A\\', ['A\\'] = 'A\\',
	['i'] = 'e', ['I'] = 'e', ['i/'] = 'e/', ['I/'] = 'e/', ['i\\'] = 'e\\', ['I\\'] = 'e\\',
	['u'] = 'o', ['U'] = 'o', ['u/'] = 'o/', ['U/'] = 'o/', ['u\\'] = 'o\\', ['U\\'] = 'o\\',
	['e'] = 'E', ['E'] = 'E', ['e/'] = 'E/', ['E/'] = 'E/', ['e\\'] = 'E\\', ['E\\'] = 'E\\',
	['o'] = 'O', ['O'] = 'O', ['o/'] = 'O/', ['O/'] = 'O/', ['o\\'] = 'O\\', ['O\\'] = 'O\\',
	['f'] = 'ar', ['F'] = 'ar', ['f/'] = 'a/r', ['F/'] = 'a/r', ['f\\'] = 'a\\r', ['F\\'] = 'a\\r',
}

export.shorten = {
	['a'] = 'a', ['A'] = 'a', ['a/'] = 'a/', ['A/'] = 'a/', ['a\\'] = 'a\\', ['A\\'] = 'a\\',
	['i'] = 'i', ['I'] = 'i', ['i/'] = 'i/', ['I/'] = 'i/', ['i\\'] = 'i\\', ['I\\'] = 'i\\',
	['u'] = 'u', ['U'] = 'u', ['u/'] = 'u/', ['U/'] = 'u/', ['u\\'] = 'u\\', ['U\\'] = 'u\\',
	['f'] = 'f', ['F'] = 'f', ['f/'] = 'f/', ['F/'] = 'f/', ['f\\'] = 'f\\', ['F\\'] = 'f\\',
}

export.lengthen = {
	['a'] = 'A', ['A'] = 'A', ['a/'] = 'A/', ['A/'] = 'A/', ['a\\'] = 'A\\', ['A\\'] = 'A\\',
	['i'] = 'I', ['I'] = 'I', ['i/'] = 'I/', ['I/'] = 'I/', ['i\\'] = 'I\\', ['I\\'] = 'I\\',
	['u'] = 'U', ['U'] = 'U', ['u/'] = 'U/', ['U/'] = 'U/', ['u\\'] = 'U\\', ['U\\'] = 'U\\',
	['f'] = 'F', ['F'] = 'F', ['f/'] = 'F/', ['F/'] = 'F/', ['f\\'] = 'F\\', ['F\\'] = 'F\\',
}

export.split_diphthong = {
	['e'] = 'ay', ['e/'] = 'a/y', ['e\\'] = 'a\\y',
	['E'] = 'Ay', ['E/'] = 'A/y', ['E\\'] = 'A\\y',
	['o'] = 'av', ['o/'] = 'a/v', ['o\\'] = 'a\\v',
	['O'] = 'Av', ['O/'] = 'A/v', ['O\\'] = 'A\\v',
}

export.semivowel_to_cons = {
	['i'] = 'y', ['I'] = 'y',
	['u'] = 'v', ['U'] = 'v',
	['f'] = 'r', ['F'] = 'r',
	['x'] = 'l', ['X'] = 'l',
}

local insert_glide = {
	['i'] = 'iy', ['I'] = 'iy', ['i/'] = 'i/y', ['I/'] = 'i/y', ['i\\'] = 'i\\y', ['I\\'] = 'i\\y',
	['u'] = 'uv', ['U'] = 'uv', ['u/'] = 'u/v', ['U/'] = 'u/v', ['u\\'] = 'u\\v', ['U\\'] = 'u\\v',
}

local to_final = {
	['k'] = 'k', ['K'] = 'k', ['g'] = 'k', ['G'] = 'k',
	['w'] = 'w', ['W'] = 'w', ['q'] = 'w', ['Q'] = 'w',
	['t'] = 't', ['T'] = 't', ['d'] = 't', ['D'] = 't',
	['p'] = 'p', ['P'] = 'p', ['b'] = 'p', ['B'] = 'p',
	['Y'] = 'N',
}

local dental_to_retroflex = {
	['t'] = 'w', ['T'] = 'W', ['d'] = 'q', ['Q'] = 'Q', ['n'] = 'R',
}

local deaspirate = {
	['K'] = 'k', ['G'] = 'g',
	['C'] = 'c', ['J'] = 'j',
	['W'] = 'w', ['Q'] = 'q',
	['T'] = 't', ['D'] = 'd',
	['P'] = 'p', ['B'] = 'b',
	['h'] = 'g',
}

function export.is_monosyllabic(text)
	return match(text, "^" .. export.consonant .. "*" .. export.vowel .. export.accent .. "?" .. export.consonant .. "*$")
end

local function absolute_final(text, ambig_hint)
	if ends_with(text, export.consonant .. export.consonant) then -- at least 2 consonants
		-- take the first of the cluster
		text = gsub(text, "(" .. export.consonant .. "+)$",
			function(cluster) return sub(cluster, 1, 1) end)
	end
	-- L, v, and y are not handled as they should not appear finally
	if ends_with(text, "[kwtpNRnmlaAiIuUeEoOfFxXH][/\\]?") then
		-- do nothing
	elseif ends_with(text, "M") then -- just in case
		text = gsub(text, ".$", "m")
	elseif ends_with(text, "[sr]") then
		text = gsub(text, ".$", "H")
	elseif ends_with(text, "[KgGWqQTdDPbBY]") then
		text = gsub(text, ".$", to_final)
	elseif ends_with(text, "[cCjJhSz]") then
		text = gsub(text, ".$", ambig_hint)
	end
	return text
end

function export.retroflexion(stem, ending)
	if ends_with(stem, "[iIeEfFxuUoOrk][/\\]?[HM]?") then
		ending = gsub(ending, "^s([^rfF])", "z%1")
	end
	if ends_with(stem, "[iIeEfFxuUoOrk][/\\]?[HM]?s") and starts_with(ending, "[^rfF]") then
		stem = gsub(stem, "s$", "z")
	end
	if ends_with(stem, "z") then
		ending = gsub(ending, "^[tTdDn]*", function(dentals) return gsub(dentals, ".", dental_to_retroflex) end)
	end
	if ends_with(stem, "[zrfF][^cCjJYwWqQRtTdDnSsl]*") then
		ending = gsub(ending,
			"^([^cCjJYwWqQRtTdDnSsl]*)n([aAiIeEfFxuUoOynmv])",
			function(pre, post)
				return pre .. "R" .. post
			end)
	end
	if ends_with(stem, "[zrfF][^cCjJYwWqQRtTdDnSsl]*n") and starts_with(ending, "[aAiIeEfFxuUoOynmv]") then
		stem = gsub(stem, "n$", "R")
	end
	-- for safety
	ending = gsub(ending,
		"([zrfF][^cCjJYwWqQRtTdDnSsl]*)n([aAiIeEfFxuUoOynmv])",
		function(pre, post)
			return pre .. "R" .. post
		end)
	return stem, ending
end

local function combine_accent(stem, ending, has_accent, accent_override, mono, recessive)
	if has_accent then
		if recessive then
			local combined = stem .. ending
			combined = gsub(combined, export.accent, "") -- remove any accent
			combined = gsub(combined, "^([^" .. export.vowel_list .. "]-)(" .. export.vowel .. ")", "%1%2/")
			return combined
		elseif accent_override then
			stem = gsub(stem, export.accent, "")
		elseif mono and match(ending, export.accent) then
			stem = gsub(stem, export.accent, "")
		elseif match(stem, export.accent) and match(ending, export.accent) then
			ending = gsub(ending, export.accent, "")
		end
	end
	return stem .. ending
end

function export.internal_sandhi(input_table)
	local stem, ending = input_table.stem, input_table.ending
	local last, acc, first, combined
	-- explicitly ignored are CV, C + semivowel, or C + nasal
	if ending == "" then
		return absolute_final(stem, input_table.ambig_hint)
	elseif starts_with(ending, export.vowel) then -- ending starts with vowel
		if ends_with(stem, export.vowel_with_accent) then -- stem ends with vowel
			-- strip last vowel and accent off stem
			stem, last, acc = match(stem, "^(.*)(" .. export.vowel .. ")(" .. export.accent .. "?)$")
			-- strip first vowel off ending
			first, ending = match(ending, "^(.)(.*)$")
			if match(last, '[iIuU]') and input_table.mono then
				stem = stem .. insert_glide[last .. acc]
				ending = first .. ending
			elseif lower(last) == lower(first) then -- homorganic
				ending = upper(first) .. acc .. ending
			elseif lower(last) == "a" then -- gunation and vrddhization
				ending = export.up_one_grade[first .. acc] .. ending
			elseif export.semivowel_to_cons[last] then
				stem = stem .. export.semivowel_to_cons[last]
				ending = first .. (acc == "/" and "\\" or "") .. ending
			elseif export.split_diphthong[last] then -- guna and vrddhi splitting
				stem = stem .. export.split_diphthong[last .. acc]
				ending = first .. ending
			end
		end
		-- all consonants unchanged
	elseif ends_with(stem, "[iu][/\\]?[rv]") and input_table.mono then
		stem = gsub(stem, "([iu][/\\]?)([rv])$", function(vow, glide) return export.lengthen[vow] .. glide end)
	elseif ends_with(stem, "[cj]") and starts_with(ending, "n") then
		ending = gsub(ending, "^.", "Y")
	elseif ends_with(stem, "S") and starts_with(ending, "s") then
		stem = gsub(stem, ".$", "k")
	elseif ends_with(stem, "s") and starts_with(ending, "s") and input_table.mono then
		stem = gsub(stem, ".$", "t")
	elseif ends_with(stem, "s") and starts_with(ending, "B") and input_table.mono then
		stem = gsub(stem, ".$", "d")
	elseif ends_with(stem, "j") and starts_with(ending, "[tT]") and input_table.j_to_z then
		stem = gsub(stem, ".$", "z")
	elseif ends_with(stem, "h") and starts_with(ending, "[tTdD]") then
		stem = gsub(stem, "([aiu]?)([/\\]?)h$", function(vow, acc) return (export.lengthen[vow] or "") .. acc end)
		ending = gsub(ending, "[tTdD]", "Q")
	elseif ends_with(stem, "[GJQDBh]") and starts_with(ending, "[tT]") then
		stem = gsub(stem, ".$", deaspirate)
		ending = gsub(ending, "^.", "D")
	elseif ends_with(stem, export.consonant) and starts_with(ending, export.consonant) then
		if input_table.final then
			if ends_with(stem, export.consonant .. export.consonant) then -- at least 2 consonants
				-- take the first of the cluster
				stem = gsub(stem, "(" .. export.consonant .. "+)$",
					function(cluster) return sub(cluster, 1, 1) end)
			end
			if ends_with(stem, "[KgGWqQTdDPbB]") then
				stem = gsub(stem, ".$", to_final)
			elseif ends_with(stem, "[cCjJhSz]") then
				stem = gsub(stem, ".$", input_table.ambig_hint)
			end
		end
		if ends_with(stem, "[kwp]") then
			if starts_with(ending, "[gGjJqQdDbB]") then
				stem = gsub(stem, ".$", {['k'] = 'g', ['w'] = 'q', ['p'] = 'b'})
			elseif starts_with(ending, "h") then
				stem = gsub(stem, ".$", {['k'] = 'g', ['w'] = 'q', ['p'] = 'b'})
				ending = gsub(ending, "^.", gsub(stem, ".$", {['k'] = 'G', ['w'] = 'Q', ['p'] = 'B'}))
			end
		elseif ends_with(stem, "t") then
			if starts_with(ending, "[cCjJwWqQ]") then
				stem = gsub(stem, ".$", gsub(ending, "^.",
					{
						['c'] = 'c', ['C'] = 'c', ['j'] = 'j', ['J'] = 'J',
						['w'] = 'w', ['W'] = 'w', ['q'] = 'q', ['Q'] = 'q',
					}))
			elseif starts_with(ending, "S") then
				stem = gsub(stem, ".$", "c")
				ending = gsub(ending, "^.", "C")
			elseif starts_with(ending, "[gGdDbB]") then
				stem = gsub(stem, ".$", "d")
			elseif starts_with(ending, "[nm]") then
				stem = gsub(stem, ".$", "n")
			elseif starts_with(ending, "h") then
				stem = gsub(stem, ".$", "d")
				ending = gsub(ending, "^.", "D")
			end
		elseif ends_with(stem, "m") then
			if starts_with(ending, "[hSzs]") then
				stem = gsub(stem, ".$", "M")
			elseif starts_with(ending, "[^yrln]") then
				stem = gsub(stem, ".$", "n")
			end
		elseif ends_with(stem, "n") then
			if starts_with(ending, "[hSzs]") then
				stem = gsub(stem, ".$", "M")
			elseif starts_with(ending, "[kKgGcCjJwWqQpPbBl]") then
				stem = gsub(stem, ".$",
					{
						['k'] = 'N', ['K'] = 'N', ['g'] = 'N', ['G'] = 'N',
						['c'] = 'Y', ['C'] = 'Y', ['j'] = 'Y', ['J'] = 'Y',
						['w'] = 'R', ['W'] = 'R', ['q'] = 'R', ['Q'] = 'R',
						['p'] = 'm', ['P'] = 'm', ['b'] = 'm', ['B'] = 'm',
						['l'] = 'M', -- or 'l~'
					})
			end
		elseif ends_with(stem, "[aA][/\\]?[sHr]") and starts_with(ending, "[rgGdDbByvjJqQlLhnm]") then
			stem = gsub(stem, "([aA])([/\\]?)[sHr]$",
				function(vow, acc) return (vow == "a" and "o" or "A") .. acc end)
		elseif ends_with(stem, "[sHr]") then
			if starts_with(ending, "[kKpPzsS]") then
				stem = gsub(stem, ".$", "H")
			elseif starts_with(ending, "[cCwWtT]") then
				stem = gsub(stem, ".$",
					{
						['c'] = 'S', ['C'] = 'S',
						['w'] = 'z', ['W'] = 'Z',
						['t'] = 's', ['T'] = 's'
					})
			elseif starts_with(ending, "r") then
				stem = gsub(stem, "(" .. export.vowel .. "[/\\])[sHr]$", function(vow) return export.lengthen[vow] or vow end)
			elseif starts_with(ending, "[gGjJqQdDbByvlLhnm]") then
				stem = gsub(stem, ".$", "r")
			end
		end
	end
	stem, ending = export.retroflexion(stem, ending)
	combined = combine_accent(stem, ending, input_table.has_accent, input_table.accent_override, input_table.mono, input_table.recessive)
	return absolute_final(combined, input_table.ambig_hint)
end

return export