Modül:ckb-alfabeçeviri

-- Authors: JavaScript ئاسۆ; Lua Ghybu, Calak

local export = {}

local gsub = mw.ustring.gsub
local U = mw.ustring.char

local mapping = {
	 = "a",  = "b",  = "ç",  = "c",  = "d",  = "e",  = "ê",  = "f",  = "g",
	 = "h",  = "h",  = "ḧ",  = "j",  = "k",  = "ll",  = "l",  = "m",  = "n",
	 = "o",  = "p",  = "q",  = "r",  = "r",  = "s",  = "ş",  = "t",  = "v",
	 = "x",  = "ẍ",  = "z",  = "",  = "'",
	
	 = "", -- ZWNJ (zero-width non-joiner)
	 = "", -- kashida, no sound
	
	-- numerals
     = "1",  = "2",  = "3",  = "4",  = "5",
	 = "6",  = "7",  = "8",  = "9",  = "0",
	-- persian variants to numerals
	 = "1",  = "2",  = "3",  = "4",  = "5",
	 = "6",  = "7",  = "8",  = "9",  = "0",
}

	-- punctuation (leave on separate lines)
local punctuation = {
	 = "?", -- question mark
	 = ",", -- comma
	 = ";", -- semicolon
	 = '“', -- quotation mark
	 = '”', -- quotation mark
	 = "%", -- percent
	 = "‰", -- per mille
	 = ".", -- decimals
	 = ",", -- thousand
}

-- translit
local function tr_word(word)
	
    word = gsub(word, '.', punctuation)
		
	--Remove punctuation at the end of the word.
	if mw.ustring.find(word, '$') then
	   ponct =  mw.ustring.sub(word, -1)
	   word = gsub(word, '$', '')
	else
	   word = word
	   ponct = ''
	end
	
	word = gsub(word, 'ه‌', "ە") --correct unicode for letter ە
	-- U+0647 (Arabic letter heh) + U+200C (zero-width non-joiner)
	-- ↓
	-- U+06D5 (Arabic letter ae)
	
	-- diacritics
	word = gsub(word, 'ْ', "i") -- U+0652, Arabic sukun
	word = gsub(word, 'ِ', "i") -- U+0650, Arabic kasra
	 
	--managing 'و' and 'ی'
	word = gsub(word, 'و()', "w%1") --و + vowel => w (e.g. wan)
	word = gsub(word, 'ی()', "y%1") --ی + vowel => y (e.g. yas)
	word = gsub(word, '()و', "%1w") --vowel + و => w (e.g. kew)
	word = gsub(word, '()ی', "%1y") --vowel + ی => y (e.g. bey)
	word = gsub(word, '^و$', "û") --non-letter + 'و' + non-letter => û (=and)
	word = gsub(word, '()و', "%1w") --non-letter + 'و' => w (e.g. wtar)

	word = gsub(word, '^و', "w") --first 'و' => w (e.g. wtar)
	word = gsub(word, 'یو', "îw") --'ی' + 'و' => îw (e.g. nîw)
	word = gsub(word, '()یی', "%1îy") --'ی' + 'ی' => îy (e.g. kanîy)
	word = gsub(word, 'وی', "uy") --'و' + 'ی' => uy (e.g. buyn)
	word = gsub(word, 'وو', "û") --'و' + 'و' => û (e.g. bû)
	word = gsub(word, 'ی', "î")
	word = gsub(word, 'و', "u")
	word = gsub(word, 'uu', "û") --'و' + 'و' => û (e.g. bû)
	word = gsub(word, '()ڕ', "%1rr") --when 'ڕ' not at the beginning of a word => rr
	word = gsub(word, '()ئ', "%1'") --when 'ئ' not at the beginning of a word => '
	
	word = gsub(word, '.', mapping)
	
	--insert i where applicable
	word = gsub(word, 'll', "Ľ") -- temporary conversion to avoid seeing ll as 2 letters
	word = gsub(word, 'rr', "Ŕ") -- temporary conversion to avoid seeing rr as 2 letters
	
	word = gsub(word, '()()()()', "%1%2i%3%4") --e.g. grft -> grift
	word = gsub(word, '()()()()$', "%1%2%3i%4") --e.g. cejnt -> cejnit
	word = gsub(word, '()()()', "%1i%2%3") --e.g. wrd -> wird
	
	word = gsub(word, '()()()', "%1i%2%3") --e.g. prd -> pird
	word = gsub(word, '()()$', "%1i%2") --like above
	
	word = gsub(word, '()()()()', "%1%2i%3%4") --repeat the latter expression, in case skipped
	word = gsub(word, '()()()$', "%1%2i%3") --repeat the latter expression, in case skipped

	word = gsub(word, '^()()()', "%1i%2%3") --e.g. ktk -> kitk
	word = gsub(word, '^()()$', "%1i%2") --e.g. ktk -> kitk
	word = gsub(word, '()()()()', "%1%2i%3%4") --e.g. ktk -> kitk
	word = gsub(word, '()()()$', "%1%2i%3") --e.g. ktk -> kitk

	word = gsub(word, '()()$', "%1%2i") --e.g. j -> ji
	word = gsub(word, '^()$', "%1i") --e.g. j -> ji	


	--word = gsub(word, '()()()', "%1%2i%3") --e.g. bra -> bira
	--word = gsub(word, '^()()', "%1i%2") --e.g. bra -> bira

	--word = gsub(word, '()()', "%1i%2") --e.g. aşkra -> aşkira
	
	--word = gsub(word, 'si()', "s%1") -- sp, st cluster
	
	word = gsub(word, 'Ľ', "ll") --revert the temporary conversion
	word = gsub(word, 'Ŕ', "rr") --revert the temporary conversion

    -- Add the punctuation who had previously deleted.
    word = word .. ponct
    
	return word
end

function export.tr(text, lang, sc)
    local textTab = {}
    
    -- Create a word table separated by a space (%s).
	for _, word in ipairs(mw.text.split(text, '%s+')) do
		table.insert(textTab, word)
	end
	
	-- Tablo of translit.
	for key, word in ipairs(textTab) do
		textTab = tr_word(word)
	end

	return table.concat(textTab, ' ')
end

return export