Module:yue-pron/sandbox

local export = {}
local m_string_utils = require("Module:string utilities")

local gsub = m_string_utils.gsub
local lower = m_string_utils.lower
local split = mw.text.split

local entering_tones = {
	 = "7",  = "8",  = "9"
}

local ipa_preprocess = {
	 = {"a", "ă"},  = {"yu", "y"},  = {"ăă", "a"},  = {"uk", "ŭk"},  = {"ik", "ĭk"},
	 = {"ou", "ŏu"},  = {"eoi", "eoy"},  = {"ung", "ŭng"},  = {"ing", "ĭng"},  = {"ei", "ĕi"},
}

local ipa_initial = {
	 = "p",  = "pʰ",  = "m",  = "f", 
	 = "t",  = "tʰ",  = "n",  = "l", 
	 = "k",  = "kʰ",  = "ŋ",  = "kʷ",  = "kʷʰ",
	 = "t͡ʃ",  = "t͡ʃʰ",  = "ʃ",
	 = "t͡s",  = "t͡sʰ",  = "s", 
	 = "h",  = "w",  = "j", 
	 = ""
}

local ipa_nucleus = {
	 = "aː",  = "ɐ", 
	 = "ɛː",  = "e", 
	 = "iː",  = "ɪ", 
	 = "ɔː",  = "o", 
	 = "œː",  = "ɵ", 
	 = "uː",  = "ʊ", 
	 = "yː"
}

local ipa_coda = {
	 = "i̯",  = "u̯",  = "y̯", 
	 = "m",  = "n",  = "ŋ", 
	 = "p̚",  = "t̚",  = "k̚", 
	 = ""
}

local ipa_tone = {
	 = "<span style=\"cursor:help\" title=\"or 53\">⁵⁵</span>", 
	 = "³⁵", 
	 = "³³", 
	 = "<span style=\"cursor:help\" title=\"or 11\">²¹</span>", 
	 = "¹³", 
	 = "²²", 
	 = "⁵", 
	 = "³", 
	 = "²", 
	 = ""
}

local ipa_tone_sandhi = {
	 = "⁻",  = "⁻",  = ""
}

local ipa_syllabic = {
	 = "m̩",  = "ŋ̍"
}

local acute_accents = {
	 = "á",  = "é",  = "í",  = "ó",  = "ú"
}

local grave_accents = {
	 = "à",  = "è",  = "ì",  = "ò",  = "ù"
}

local macrons = {
	 = "ā",  = "ē",  = "ī",  = "ō",  = "ū"
}

local yale_final = {
	 = "沒有耶魯拼音",  = "a",
	 = "沒有耶魯拼音",  = "沒有耶魯拼音",  = "沒有耶魯拼音",  = "沒有耶魯拼音",  = "沒有耶魯拼音",
	 = "沒有耶魯拼音",  = "沒有耶魯拼音",
	 = "沒有耶魯拼音",  = "沒有耶魯拼音",
	 = "eu",  = "沒有耶魯拼音",  = "沒有耶魯拼音",  = "eung",  = "沒有耶魯拼音",  = "沒有耶魯拼音",  = "euk", 
	 = "eui",  = "eun",  = "eut",
	 = ""
}

local canton_pinyin_regex = {
	 = "y",  = "oe",  = "oey",  = "dz%1",  = "ts%1", 
	)()"]=function(a,b) return a .. entering_tones end,
	)()$"]=function(a,b) return a .. entering_tones end
}

local function yale_tone(final,b)
	if b == "4" or b == "5" or b == "6" then
		final = gsub(final, "(?g?)$", "h%1", 1)
	end
	if b == "1" then final = gsub(final, "", macrons, 1) end
	if b == "4" then final = gsub(final, "", grave_accents, 1) end
	if b == "2" or b == "5" then final = gsub(final, "", acute_accents, 1) end
	return final
end

function export.jyutping_to_ipa(text)
	if type(text) == "table" then text = text.args end

	if text:find("") then error("Invalid tone in Jyutping.") end
	if text:find("") then error("Please do not capitalize the Jyutping.") end
	if text:find("%-") then error("Please do not hyphenate the Jyutping.") end
	if text:find("") then error("Error in the Jyutping transcription.") end
	if text:find("h") then error("'zh'/'ch'/'sh' are non-valid Jyutping, use 'z'/'c'/'s' instead.") end
	if text:find("y") then error("Wrong usage of 'y' in Jyutping.") end
	if text:find("oei") then error("Invalid rime oei in Jyutping.") end
	if text:find("eong") then error("Invalid rime eong in Jyutping") end
	if text:find("eok") then error("Invalid rime eok in Jyutping") end

	text = text:lower()
	text = gsub(text, "jy()", "j%1")
	text = gsub(text, "%.%.%.", " ")
	text = gsub(gsub(text, ",", "隔"), "隔 ", ", ")
	local reading = split(text, "隔")
	
	local function retrieve_ipa1(a,b,c,d,e)
		return ipa_initial .. ipa_syllabic .. ipa_tone .. ipa_tone_sandhi .. ipa_tone
	end
	
	local function retrieve_ipa2(a,b,c,d,e,f)
		return (ipa_initial or error(("Unrecognised initial: \"%s\""):format(a))) ..
			(ipa_nucleus or error(("Unrecognised nucleus: \"%s\""):format(b))) ..
			(ipa_coda or error(("Unrecognised coda: \"%s\""):format(c))) ..
			(ipa_tone or error(("Unrecognised tone: \"%s\""):format(d))) ..
			ipa_tone_sandhi ..
			(ipa_tone or error(("Unrecognised tone: \"%s\""):format(f)))
	end
	
	local function get_entering_tone(a,b) return a .. entering_tones end
	
	for i = 1,#reading do
		reading = gsub(gsub(reading, "()", "%1 "), "", "")
		local syllable = split(reading:lower(), " ")
		if #syllable == 1 then
			require('Module:debug').track('yue-pron/'..(syllable))
		end
		for i = 1,#syllable do
			syllable = gsub(syllable, "()yu", "%1hyu")
			syllable = gsub(syllable, "()oe", "%1hoe")
			syllable = gsub(syllable, "()eo", "%1heo")
			if (not syllable:find("^??+?g???$") and not syllable:find("^h?g???$")) and syllable:find("") then
				error("Incorrect Jyutping format. Please check!")
			end
			syllable = gsub(syllable, "^(h?)(g?)()(?)(?)$", retrieve_ipa1)
			syllable = gsub(syllable, "()()", get_entering_tone)
			syllable = gsub(syllable, "()()$", get_entering_tone)
			for regex_idx,regex_pair in ipairs(ipa_preprocess) do
				syllable = gsub(syllable,regex_pair,regex_pair)
			end
			syllable = gsub(syllable, "spăn", "span")
			syllable = gsub(syllable,
				"(??)(?)(?g?)()(?)(?)",
				retrieve_ipa2)
		end
		reading = table.concat(syllable, " ")
	end
	return table.concat(reading, "/, /")
end

function export.jyutping_to_yale(text)
	if type(text) == "table" then text = text.args end

	text = gsub(text, "jy()", "j%1")
	text = gsub(text, ",", "隔"):gsub("隔 ", ", ")
	local reading = split(text, "隔")
	
	local function yale_tone2(a, b) return yale_tone((yale_final or a), b) end
	
	for i = 1,#reading do
		reading = gsub(reading, "%-", "")
		reading = gsub(reading, "()", "%1 ")
		local syllables = split(lower(reading), " ",true)
		for j = 1,#syllables do
			local text = syllables
			
			text = gsub(text, "",{ = "1",  = "3",  = "6"})
			text = gsub(text, "^m$",{ = "{{懸停|或m̀|m̄}}",  = "ḿ",  = "m",  = "m̀h",  = "ḿh",  = "mh"})
			text = gsub(text, "^ng$",{ = "{{懸停|或ǹg|n̄g}}",  = "ńg",  = "ng",  = "ǹgh",  = "ńgh",  = "ngh"})
			text = gsub(text, "jy?", "y")
			text = gsub(text, "",{ = "j",  = "ch"})
			text = gsub(text, "(???)()", yale_tone2)
			text = gsub(text, "(yu?)()", yale_tone)
			text = gsub(text, "",{ = "{{懸停|或à|ā}}",  = "{{懸停|或è|ē}}",  = "{{懸停|或ì|ī}}",  = "{{懸停|或ò|ō}}",  = "{{懸停|或ù|ū}}"})
			text = gsub(text, "{{懸停|或(+)|(+)}}", '<span style="cursor:help" title="or %1">%2</span>')
			
			syllables = text
		end
		reading = table.concat(syllables, " ")
		if reading:find("沒有耶魯拼音") then
			reading = "colloquial sounds not defined"
		end
	end
	return table.concat(reading, ", ")
end

function export.jyutping_to_cantonese_pinyin(text)
	if type(text) == "table" then text = text.args end
	
	text = gsub(text, "jy()", "j%1")
	text = gsub(gsub(text, ",", "隔"), "隔 ", ", ")
	local reading = split(text, "隔")
	for i=1,#reading do
		if reading:find("oe") then
			reading = "沒有教院式拼音"
		end
		for regex, replace in pairs(canton_pinyin_regex) do
			reading = gsub(reading, regex, replace)
		end
		if reading:find("沒有教院式拼音") then
			reading = "colloquial sounds not defined"
		end
	end
	return table.concat(reading, ", ")
end

function export.jyutping_to_guangdong(text)
	if type(text) == 'table' then text = text.args end
	
	local palatal = { ='j', ='q', ='x' }
	local function palatalize(a,b) return palatal .. b end
	
	local final = { ='b', ='g', ='d' }
	local function get_final(a,b) return final .. b end

	text = gsub(gsub(text, ",", "隔"), "隔 ", ", ")
	local reading = split(text, "隔")
	for i, item in ipairs(reading) do
		item = gsub(item, 'yu', 'ü')
		item = gsub(item, 'eoi', 'êü')
		item = gsub(item, 'j', 'y')

		item = gsub(item, 'oe', '沒有廣東拼音')
		item = gsub(item, '', 'ê')
		
		item = gsub(item, 'o', '沒有廣東拼音')
		item = gsub(item, 'e', '沒有廣東拼音')
		item = gsub(item, 'e', 'é')
		
		item = gsub(item, 'u', '沒有廣東拼音')

		item = gsub(item, '()()',  palatalize)

		item = gsub(item, '()ü', '%1u')

		item = gsub(item, 'a+', { ='a', ='e' } )

		item = gsub(item, '()w', '%1u')
		item = gsub(item, '()u', '%1o')

		item = gsub(item, '()(%d)', get_final)
		
		item = gsub(item, 'e', '沒有廣東拼音')
		
		if item:find('沒有廣東拼音') then
			item = 'colloquial sounds not defined'
		end
		reading = item
	end
	return table.concat(reading, ", ")
end

function export.jyutping_format(text)
	if type(text) == "table" then text = text.args end
	
	if text:find("%[%[") then
		return "just a lengthy text to ensure it works"
	end
	
	text = split(gsub(gsub(text, ",", "隔"), "隔 ", ", "), "隔")
	for i, to_be_processed in ipairs(text) do
		text = gsub(to_be_processed, "%-()", "%1")
	end
	
	return text
end

function export.hoisanva_to_ipa(text)
	local hsv_initial = {
		 = "p",  = "pʰ",  = "ᵐb",  = "f",  = "v",
		 = "t",  = "tʰ",  = "ⁿd",  = "l",  = "ɬ",
		 = "k",  = "kʰ",  = "ᵑɡ", 
		 = "t͡s",  = "t͡sʰ",
		 = "j",  = "s",  = "h",  = ""
	}
	
	local hsv_final = {
		 = "a",  = "ai",  = "au",  = "am",
		 = "an",  = "aŋ",  = "ap̚",  = "at̚",
		 = "ak̚", 

		 = "i",  = "iu",  = "im",  = "in", 
		 = "ip̚",  = "it̚",

		 = "iɛ",  = "iau",  = "iam",  = "iaŋ", 
		 = "iap̚",  = "iak̚",

		 = "u",  = "ui",  = "un",  = "ut̚", 

		 = "ə",  = "ei",  = "eu",  = "em",  = "en", 
		 = "ɵŋ",  = "ep̚",  = "et̚",  = "ɵk̚",  = "ɵt̚",

		 = "ᵘɔ",  = "ᵘɔi",  = "ᵘɔn",  = "ɔŋ", 
		 = "ᵘɔt̚",  = "ɔk̚", 
		
		 = "m̩"
	}
	
	local hsv_tone = { "³³", "⁵⁵", "²²", "²¹", "³²" }
	
	local gsplit = mw.text.gsplit

	local result = {}
	for word in gsplit(text, ",") do
		local initial, final, tone, tone_ch, word_result = "", "", "", "", {}
		for syllable in gsplit(word, " ") do
			initial, final, tone, tone_ch = syllable:match("^(*)(*)()(??%*?)$")
			if final == "" then final, initial = initial, "" end
			
			if not hsv_initial or not hsv_final or not hsv_tone or (tone == "2" and tone_ch == "*") then
				error("Syllable '" .. syllable .. "' is not a valid syllable for IPA conversion.")
			end
			
			table.insert(word_result,
				hsv_initial .. hsv_final .. hsv_tone ..
				(tone_ch ~= "" and "⁻" or "") ..
				(tone_ch == "*" and hsv_tone or (tone_ch:find("^%-") and hsv_tone or "")) ..
				(tone_ch:sub(-1, -1) == "*" and "⁵" or ""))
		end
		table.insert(result, table.concat(word_result, " "))
	end
	return "/" .. table.concat(result, "/, /") .. "/"
end

function export.jyutping_headword(frame)
	local params = {
		 = {},
	}
	
	local args = require("Module:parameters").process(frame:getParent().args, params, nil, "yue-pron", "jyutping_headword")
	
	local head = args.head or mw.title.getCurrentTitle().text
	
	local yue = require("Module:languages").getByCode("yue")
	local Latn = require("Module:scripts").getByCode("Latn")
	
	head = head:gsub("%d?%d?", "<sup>%0</sup>")
	
	return require("Module:headword").full_headword{lang = yue, sc = Latn, heads = {head}, pos_category = "jyutping"}
end

return export