Module:User:Chuterix/sandbox/kyoto-pron

local export = {} local gsub = mw.ustring.gsub local match = mw.ustring.match local sub = mw.ustring.sub local len = mw.ustring.len

local lang = require("Module:languages").getByCode("ja") local kana_to_romaji = require("Module:Hrkt-translit").tr local m_accent = require("Module:accent qualifier") -- also Module:qualifier

local PAGENAME = mw.title.getCurrentTitle.text

local function quote(text) return "“" .. text .. "”" end

local ref_template_name_data = { ['DJR'] = 'R:Daijirin', ['DJS'] = 'R:Daijisen', ['KDJ'] = 'R:Kokugo Dai Jiten', ['NHK'] = 'R:NHK Hatsuon', ['SMK2'] = 'R:Shinmeikai2', ['SMK5'] = 'R:Shinmeikai5', ['SMK7'] = 'R:Shinmeikai7', ['ZAJ'] = 'R:Zenkoku Akusento Jiten', ['JEL'] = 'R:Kenkyusha JEL Pocket', ['JAC'] = 'R:ja:JAccent', }

local function add_acc_refs(text) local output = {} for ref_name in mw.text.gsplit(text, ',') do		mw.log(ref_name) ref_template_name = ref_template_name_data[ref_name] if ref_template_name then table.insert(output, mw.getCurrentFrame:extensionTag{				name = 'ref',				args = { name = ref_name },				content = '',			}) elseif mw.ustring.match(ref_name, 'ref') then table.insert(output, mw.getCurrentFrame:preprocess(ref_name)) else -- Special:WhatLinksHere/Wiktionary:Tracking/ja-pron/unrecognized ref require("Module:debug").track("ja-pron/unrecognized ref") end end return table.concat(output) end

function export.show(frame) local params = { [1] = {default = PAGENAME, list = true}, ["accent"] = {list = true}, ["accent=_loc"] = {list = true}, ["accent=_ref"] = {list = true, allow_holes = true}, ["accent=_note"] = {list = true, allow_holes = true}, ["acc"] = {alias_of = "accent", list = true}, ["acc=_loc"] = {alias_of = "accent_loc", list = true},	-- using "accent=_loc" (etc.) doesn't work ["acc=_ref"] = {alias_of = "accent_ref", list = true}, ["acc=_note"] = {alias_of = "accent_note", list = true}, ["dev"] = {}, ["dev2"] = {}, ["devm"] = {}, ["noipa"] = {}, ["a"] = {alias_of = "audio"}, ["audio"] = {} }	local args = require("Module:parameters").process(frame:getParent.args, params) local au = args.audio local dev = args.dev or args.devm local dev2 = args.dev2 local maxindex = table.getn(args[1]) local html_list_main = mw.html.create('ul') local text -- Deals with the accents local a, al, ar, an = args.accent, args.accent_loc, args.accent_ref, args.accent_note local no_acc = true for i, position in ipairs(a) do		local result no_acc = false

text = args[1][math.min(maxindex,i)] if not al[i] then al[i] = "Kyoto" end result = m_accent.format_qualifiers(lang, {al[i]}) .. " "		result = result .. export.accent(text, position, dev, dev2) if ar[i] then result = result .. add_acc_refs(ar[i]) else require("Module:debug").track("ja-pron/unsourced accent") end result = result .. (an[i] and (" " .. an[i]) or "")

html_list_main:tag('li'):wikitext(			result		) end if no_acc then -- track when entries have no pitch information require("Module:debug").track("ja-pron/no accent") end -- Deals with the IPA if not noipa then local m_IPA = require("Module:IPA") for i, text in ipairs(args[1]) do			local sortkey = (lang:makeSortKey(text)) html_list_main:tag('li'):wikitext(				m_IPA.format_IPA_full {					lang = lang,					items = ,					sort_key = sortkey,				}			) end end -- Deals with the audio if au then sortkey = (lang:makeSortKey(args[1][1])) html_list_main:tag('li'):wikitext(			' '		) end return '\n' .. tostring(html_list_main) end

function export.ipa(text, dev, dev2) if type(text) == "table" then text, dev, dev2 = text.args[1], text.args["dev"], text.args["dev2"] end dev = dev or "" dev2 = dev2 or "" if dev2 ~= "" then error('Please remove parameter dev2 and change parameter dev to \"dev=' .. dev .. ',' .. dev2 .. '"') end local position_mora = {} for i=1,len(text) do		if not match(sub(text,i,i), "[ ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒%.]") then if sub(text,i+1,i+1) and match(sub(text,i+1,i+1), "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒]") then table.insert(position_mora, i+1) else table.insert(position_mora, i)			end end end -- insert @ to stand for devoicing if dev ~= "" then for position in mw.text.gsplit(dev,",") do			position = tonumber(position) if #position_mora == position then text = text .. "@"			else local position_devspace = position_mora[position] text = sub(text, 1, position_devspace) .. "@" .. sub(text, position_devspace+1, -1) end for i=position+1,#position_mora do				position_mora[i] = position_mora[i] + 1 end end end text = kana_to_romaji(text, "ja", nil, {keep_period = true}) text = gsub(text, "@", "̥") text = gsub(text, "ー", "ː") text = gsub(text, "'", "ʔ") text = gsub(text, "[ptkbjgzsdr][ptckbjgzsdr][hs]?", {		["pp"] = "p̚p", ["tch"] = "t̚ch", ["kk"] = "k̚k", ["bb"] = "b̚b̥", ["jj"] = "d̚j",		["dd"] = "d̚d̥", ["gg"] = "g̚g̊", ["zz"] = "d̚z", ["tt"] = "t̚t", ["tts"] = "t̚ts",		["rr"] = "r̚r", ["ssh"] = "ɕː" }) text = gsub(text, "ei", "ē") text = gsub(text, "[āēīōūfvjryz]", {		["ā"] = "aː", ["ē"] = "eː", ["ī"] = "iː", ["ō"] = "oː", ["ū"] = "uː", 		["f"] = "ɸ", ["v"] = "b", ["j"] = "d͡ʑ", ["r"] = "ɾ", ["y"] = "j", ["z"] = "d͡z" }) text = gsub(text, "[sct][hs]", {		["sh"] = "ɕ", 		["ch"] = "t͡ɕ", 		["ts"] = "t͡s" }) text = gsub(text, "([aeiouː̥])d͡([zʑ])", "%1%2") text = gsub(text, "([pbtdkgnmɸszɾ][̥̊]?)i", "%1ʲi") text = gsub(text, "([pbtdkgnmɸszɾ][̥̊]?)j", "%1ʲ") text = gsub(text, "nʲ", "ɲ̟") text = gsub(text, "([^ ː])(ː?)n([^aeou])", "%1̃%2n%3") for i, args in pairs{ { "(ː?)n$", "̃%1ɴ" }, { "n( ?)([pbm])", "m%1%2" }, { "n( ?)(.͡[ɕʑ])", "ɲ̟%1%2" }, { "n( ?)ɲ̟", "ɲ̟%1ɲ̟" }, { "n( ?)([kg])(ʲ?)", "ŋ%1%3%2%3" }, { "n( ?)([ɸszɕhjw])", "ɰ̃%1%2" }, { "nʔ", "ɰ̃" }, { "n ([aeiou])", "ɰ̃ %1" }, } do		text = gsub(text, args[1], args[2]) end text = gsub(text, "h[iju]", {		["hi"] = "çi", ["hj"] = "ç", 		["hu"] = "ɸu" }) text = gsub(text, "h([çɸ])", "%1%1") text = gsub(text, "([snhçɸmɾjw])%1", "%1ː") text = gsub(text, "ːʲ", "ʲː") text = gsub(text, "̚(.[̥̊]?)ʲ", "̚ʲ%1ʲ") text = gsub(text, "[aeiouw]", {		["a"] = "a̠", 		["e"] = "e̞", 		["o"] = "o̞", 		["u"] = "ɯ̟ᵝ", 		["w"] = "ɰᵝ"}) text = gsub(text, "([szɕʑɲçʲ])ɯ̟", "%1ɨ") text = gsub(text, "ᵝ̥", "̥ᵝ") text = gsub(text, "ᵝ̃", "̃ᵝ") text = gsub(text, "̠[̥̃][̥̃]", "̥̃˗") text = gsub(text, "̞[̥̃][̥̃]", "̥̃˕") text = gsub(text, "̟[̥̃][̥̃]", "̥̃˖") text = gsub(text, "([̠̞̟])̥", "%1̊") text = gsub(text, "%.", "") text = gsub(text, "'", ".") text = gsub(text, "g", "ɡ") return text end

function export.rise_and_fall(word, rftype) word = gsub(word, "([おこごそぞとどのほぼぽもよろぉょオコゴソゾトドノホボポモヨロォョ])([うウ])", "%1.%2") word = gsub(word, "([えけげせぜてでねへべぺめれゑぇエケゲセゼテデネヘベペメレェ])([いイ])", "%1.%2") word = kana_to_romaji(word, "ja")

if rftype == "rise" then word = gsub(word, ".", {			["a"] = "á", ["e"] = "é", ["i"] = "í", ["o"] = "ó", ["u"] = "ú", 			["ā"] = "áá", ["ē"] = "éé", ["ī"] = "íí", ["ō"] = "óó", ["ū"] = "úú" }) word = gsub(gsub(word, "n([bcdfghjkmnprstvw%'z ])", "ń%1"), "n$", "ń") elseif rftype == "fall" then word = gsub(word, ".", {			["a"] = "à", ["e"] = "è", ["i"] = "ì", ["o"] = "ò", ["u"] = "ù", 			["ā"] = "àà", ["ē"] = "èè", ["ī"] = "ìì", ["ō"] = "òò", ["ū"] = "ùù" }) word = gsub(gsub(word, "n([bcdfghjkmnprstvw%'z ])", "ǹ%1"), "n$", "ǹ") else return error("Type not recognised.") end return word end

-- Module:ja-ojad and Module:ja-infl-demo rely on the output format of this function function export.accent(text, class, dev, dev2) local result

if(type(text)) == "table" then text, class, dev, dev2 = text.args[1], text.args[2], text.args["dev"], text.args["dev2"] end text = gsub(text, "([おこごそぞとどのほぼぽもよろぉょオコゴソゾトドノホボポモヨロォョ])[うウ]", "%1ー") text = gsub(text, "([えけげせぜてでねへべぺめれゑぇエケゲセゼテデネヘベペメレェ])[いイ]", "%1ー") text = gsub(text, "%.", "") if dev == "" then dev = false end if dev2 == "" then dev2 = false end local down_first = "" local down_last = "&#8203; " local high_first = "" local start = "" local romaji_start = "  [" local romaji_last = "] " local last = " " local position_kana = {}      --position of each kana (ぁ counted), text without space local position_mora = {}      --position of each mora (ぁ not counted), text without space local position_mora_space = {} --position of each mora (ぁ not counted), text with space for i=1,mw.ustring.len(text) do		if not mw.ustring.match(sub(text,i,i), "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒 ]") then local extra = mw.ustring.len(mw.ustring.match(sub(text,i+1), "^[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒]*")) table.insert(position_mora_space, i+extra) end end local space_removed = mw.ustring.gsub(text," ","") for i=1,mw.ustring.len(space_removed) do		table.insert(position_kana, i)		if not mw.ustring.match(sub(space_removed,i,i), "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒]") then local extra = mw.ustring.len(mw.ustring.match(sub(space_removed,i+1), "^[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒]*")) table.insert(position_mora, i+extra) end end local acc_type, acc_number if match(class, "^[h0]$") then acc_type, acc_number = "h", 0 elseif match(class, "^[a1]$") then acc_type, acc_number = "a", 1 elseif match(class, "^o$") then acc_type = "o" acc_number = len(gsub(text, "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒 ]", "")) end if match(class, "^[0-9]+$") and not match(class,"^[01]$") then class = gsub(class, "[on]", "") acc_number = tonumber(class) local morae_count = len(gsub(text, "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒 ]", "")) if morae_count == acc_number then acc_type = "o" elseif morae_count < acc_number then return error(("Mora count (%d) is smaller than position of downstep mora (%d).")				:format(morae_count, acc_number)) else acc_type = "n" end elseif not acc_number then acc_number = class end local start_index = 1 while match(sub(text, start_index+1, start_index+1), "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒]") do		start_index = start_index + 1 end local kanas = {} local single_mora for i=1,mw.ustring.len(text) do		if not mw.ustring.match(sub(text,i,i), "[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒 ]") then single_mora = gsub(sub(text, i, -1), "^(.[ァィゥェォャュョヮぁぃぅぇぉゃゅょゎ𛅑𛅐𛅒]*).*", "%1") table.insert(kanas, single_mora) end end local function kana_devoice(text) return ' ' .. text .. " "	end if dev then for position in mw.text.gsplit(dev, ",") do			position = tonumber(position) kanas[position] = kana_devoice(kanas[position]) end end romaji_text = gsub(text, "([おこそとのほもよろをごぞどぼぽょぉオコソトノホモヨロヲゴゾドボポョォ])ー", "%1お") romaji_text = gsub(romaji_text, "([えけせてねへめれゑげぜでべぺぇエケセテネヘメレヱゲゼデベペェ])ー", "%1え") romaji_text = gsub(romaji_text, "([うくすつぬふむゆるぐずづぶぷゅうウクスツヌフムユルグズヅブプュゥゔヴ])ー", "%1う") romaji_text = gsub(romaji_text, "([いきしちにひみりゐぎじぢびぴぃイキシチニヒミリヰギジヂビピィ])ー", "%1い") romaji_text = gsub(romaji_text, "([あかさたなはまやらわんがざだばぱゃぁアカサタナハマヤラワンガザダバパャァ])ー", "%1あ") local romajis = mw.text.split(romaji_text, "") local function count_nspaces(text, index) local i, sample, nspaces = 0, "", 0 while len(sample) < index do			i = i + 1 sample, nspaces = gsub(sub(text, 1, i), " ", "") end return nspaces end local function romaji_devoice(text) -- use @ instead of ̥ return text .. "@"	end if dev then for position in mw.text.gsplit(dev,",") do			position = position_mora_space[tonumber(position)] romajis[position] = romaji_devoice(romajis[position]) end end if acc_type == "n" then r_start_index = start_index + count_nspaces(romaji_text, start_index) local r_index = position_mora_space[acc_number] local k_index = acc_number r_parts = { [1] = table.concat(romajis, "", 1, r_start_index), [2] = table.concat(romajis, "", r_start_index + 1, r_index), [3] = table.concat(romajis, "", r_index + 1, #romajis) }		k_parts = { [1] = table.concat(kanas, "", 1, 1), [2] = table.concat(kanas, "", 2, k_index), [3] = table.concat(kanas, "", k_index + 1, #kanas) }		local space2 = "" local space3 = "" if sub(r_parts[2], 1, 1) == " " then space2 = " " end if sub(r_parts[3], 1, 1) == " " then space3 = " " end result = start .. k_parts[1] .. down_first .. k_parts[2] .. down_last .. k_parts[3] .. last .. romaji_start .. export.rise_and_fall(r_parts[1], "fall") .. space2 .. export.rise_and_fall(r_parts[2], "rise") .. "ꜜ" .. space3 .. export.rise_and_fall(r_parts[3], "fall") .. romaji_last .. "(Nakadaka – [" .. acc_number .. "])" else r_start_index = start_index + count_nspaces(romaji_text, start_index) r_parts = { [1] = table.concat(romajis, "", 1, r_start_index), [2] = table.concat(romajis, "", r_start_index + 1, #romajis) }		k_parts = { [1] = table.concat(kanas, "", 1, 1), [2] = table.concat(kanas, "", 2, #kanas) }		local space2 = "" if sub(r_parts[2], 1, 1) == " " then space2 = " " end if acc_type == "h" then result = start .. k_parts[1] .. last .. k_parts[2] .. last .. last .. romaji_start .. export.rise_and_fall(r_parts[1], "fall") .. space2 .. export.rise_and_fall(r_parts[2], "rise") .. romaji_last .. "(Heiban – [" .. acc_number .. "])" elseif acc_type == "a" then result = start .. down_first .. k_parts[1] .. down_last .. k_parts[2] .. last .. romaji_start .. export.rise_and_fall(r_parts[1], "rise") .. "ꜜ" .. space2 .. export.rise_and_fall(r_parts[2], "fall") .. romaji_last .. "(Atamadaka – [" .. acc_number .. "])" elseif acc_type == "o" then result = start .. k_parts[1] .. down_first .. k_parts[2] .. down_last .. last .. romaji_start .. export.rise_and_fall(r_parts[1], "fall") .. space2 .. export.rise_and_fall(r_parts[2], "rise") .. "ꜜ" .. romaji_last .. "(Odaka – [" .. acc_number .. "])" else return error("Accent type not recognised.") end end result = gsub(result, "(.)@", " %1 ") return result end

return export