Modül:ja-kanjitab

local export = {}

local m_utilities = require("Modül:araçlar")
local m_ja = require("Modül:ja")
--[=[
	Other modules used: ], ], ]
]=]

local title = mw.title.getCurrentTitle()
local PAGENAME = title.text
local NAMESPACE = title.nsText

local lang = require("Modül:diller").getirKodaGore("ja")

local kanji_pattern = "㐀-䶵一-鿌\239\164\128-\239\171\153𠀀-𯨟"

local kanji_grade_links = {
	"]",
	"]",
	"]",
	"]",
	"]",
	"]",
	"]",		-- 7
	"]",	-- 8
	"]"		-- 9
}

local function track(code)
	if type(code) == "string" then
		--require("Modül:debug").track("ja-kanjitab/" .. code)
	end
end

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

-- this is the function that is called from templates
function export.show(frame)
	local params = {
		 = { list = true, allow_holes = true },
		k = { list = true, allow_holes = true },
		o = { list = true, allow_holes = true },
		r = {},
		sort = {},
		yomi = {},
		ateji = {},
		alt = {},
		pagename = {},
	}
	local args, unrecognized_args = require("Modül:parameters").process(frame:getParent().args, params, true)

	if args.pagename and NAMESPACE == "" then
		track("pagename param in mainspace")
	end
	local pagename = args.pagename or PAGENAME

	for key, value in pairs(unrecognized_args) do
		local additional
		if mw.ustring.sub(key, 1, 1) == "y" then
			additional =  " Perhaps you meant " .. quote("yomi") .. "?"
		end
		error(quote(key) .. " is not a recognized parameter." .. (additional or ""))
	end

	local categories = {}
	local cells = {}
	-- replace e.g. 時々 with 時時
	local kanji = mw.ustring.gsub(pagename, '()々', '%1%1')
	-- remove non-kanji characters
	kanji = mw.ustring.gsub(kanji, '', '')

	local kanji_length = mw.ustring.len(kanji)
	local colspan = ""
	if kanji_length > 1 then
		colspan = 'colspan="' .. kanji_length .. '" '
	end

	local table_head = [=[
{| class="wikitable kanji-table" style="text-align: center; font-size: small; float: right;"
! ]=] .. colspan .. ] karakterleri
|- lang="ja" class="Jpan" style="font-size: 2em; background: white; line-height: 1em;"

]=]

	local yomi

	-- on/kun is jūbakoyomi; NOTE: these are only applicable for two-kanji compounds
	-- kun/on is yutōyomi; NOTE: these are only applicable for two-kanji compounds
	if args.yomi then
		yomi = {}
		local extended_yomi_code = {
			o = 'on',			on = 'on',
			kanon = 'kanon',  -- kan is kan'yoon, and ko is kun+on for backward compatibility
			goon = 'goon',
			toon = 'toon',
			kan = 'kanyoon',	kanyo = 'kanyoon',		kanyoon = 'kanyoon',
			k = 'kun',			kun = 'kun',
			juku = 'jukujikun',	jukuji = 'jukujikun',	jukujikun = 'jukujikun',
			ok = "jūbakoyomi",	j = "jūbakoyomi",
			ko = "yutōyomi",	y = "yutōyomi",			yu = "yutōyomi",
			i = 'düzensiz',	irr = 'düzensiz',		irreg = 'düzensiz',	irregular = 'düzensiz',
			n = 'nanori',	nanori = 'nanori',
			 = '',			none = '',
		}
		for i in mw.text.gsplit(args.yomi, ',') do
			local _, _, a, b = mw.ustring.find(i, '^(*)(*)$')
			a = extended_yomi_code or error("The yomi type “" .. args.yomi .. "” is not recognized.")
			b = tonumber(b) or 1
			table.insert(yomi, { a, b })
		end
		if #yomi == 1 and kanji_length > 1 then
			yomi = kanji_length
		end
	else
		-- ]
		track("no yomi")
	end

	-- ]
	if yomi and (yomi == "jūbakoyomi" or yomi == "yutōyomi") and kanji_length ~= 2 then
		track("ja-pron/incorrect yutou or juubako")
	end


	-- readings and okurigana are appended to this
	local sortkey = ""

	-- ]
	-- ]
	if args.k.maxindex and args.k.maxindex > args.maxindex then
		track("too many k")
	end

	if args.o.maxindex and args.o.maxindex > args.maxindex then
		track("too many o")
	end

	local yomi_type_by_kanji = {}
	if yomi then
		for i = 1, #yomi do
			for j = 1, yomi do
				table.insert(yomi_type_by_kanji, yomi)
			end
		end
	else
		for i = 1, kanji_length do
			table.insert(yomi_type_by_kanji, '')
		end
	end

	local is_ateji = {}
	if args.ateji then
		local ateji = args.ateji
		local cat_ateji = false
		if ateji == 'y' then
			for i = 1, kanji_length do
				is_ateji = true
			end
			cat_ateji = true
		else
			for i in mw.text.gsplit(ateji, ';') do
				string.gsub(i, '^(+)$', function(a)
					is_ateji = true
					cat_ateji = true
				end)
				string.gsub(i, '^(+),(+)$', function (a, b)
					for j = tonumber(a), tonumber(b) do
						is_ateji = true
					end
					cat_ateji = true
				end)
			end
		end
		if cat_ateji then table.insert(categories, "Japonca ateji karakterini içeren sözcükler") end
	end

	-- if hiragana readings were passed,
	-- make the "spelled with ..." categories, the readings cells on the lower level and build the sort key
	-- otherwise rely on the pagename to make the original kanjitab and categories
	local cells_above = {}
	local cells_below = {}
	local kanji_pos = 1
	for i = 1, kanji_length do
		local reading = args
		local reading_kana, reading_length = nil, nil
		local cell = {}

		if reading then _, _, reading_kana, reading_length = mw.ustring.find(reading, '^(*)(*)$') end
		reading_kana = reading_kana ~= "" and reading_kana or nil
		reading_length = reading_kana and tonumber(reading_length) or 1

		if reading_length <= 1 then
			table.insert(cell, '| rowspan="2" | ')
		else
			table.insert(cell, '| colspan ="' .. reading_length .. '" | ')
		end

		-- display reading, actual reading and okurigana
		if reading_kana then
			if mw.ustring.find(reading_kana, '') and not mw.ustring.find(reading_kana, '^+$') then
					--]
					track('not all hiragana')
			end

			local actual_reading = args.k
			local okurigana = args.o

			sortkey = sortkey .. (actual_reading or reading_kana) .. (okurigana or "")

			local okurigana_text = okurigana and "(" .. okurigana .. ")" or ""
			local actual_reading_text = actual_reading and " > " .. actual_reading .. okurigana_text or ""
			local text = reading_kana .. okurigana_text .. actual_reading_text

			table.insert(cell, '<span class="Jpan" lang="ja">' .. text .. '</span>')
			if reading_length <= 1 then table.insert(cell, '<br/>') end
		end

		-- display kanji grade, categorize
		for j = kanji_pos, kanji_pos + reading_length - 1 do
			local single_kanji = mw.ustring.sub(kanji, j, j)
			local kanji_grade = m_ja.kanji_grade(single_kanji)
			local ateji_text = is_ateji and '<br/><small>(])</small>' or ''

			if reading_kana then
				-- subcategorize by reading if this is joyo kanji, not doing that for less common kanji, with exceptions
				if (kanji_grade < 8 or (
					'厭昌之芽昌浩智晃淳敦聡晃旭亮糊桂隘阿唖撫鼠阿耘迂寅已伊餡姦闊礙碍凱亥价謳嘔齧日臣桶抉兎鵜卯綾飴焙肋鮫頚糞軋烏痒捷辰叩橙揃嶋澤菱彦囃覗呑之乃鼠做寅樋堤槌机杖頼辿哉叢狢峯巳卍鱒仄他惚弘宏燕倦經痙圭禽僑鋸醵墟屹綺几翫癌劫膠昂鹸牽喧餐鑽瑣些渾梱坤國壕誦哨蒐杓爾梓荼楕躁綜楚闡閃撰專泄藉棲錘錐祷盪淘點顛填擲擢闖厨蛋潭腿冪碧劈焚祓弗憑誹砒婢挽拔撥剥胚播乃狼牢蓮礫醂龍榴蕾酉祐佑耶也蔓曼沫邁呆硼牡甫步矮狸苔'
				):find(single_kanji)) and yomi_type_by_kanji ~= 'düzensiz' and yomi_type_by_kanji ~= 'jukujikun' and reading_length == 1 then
					table.insert(categories, "Japonca " .. reading_kana .. " okunuşlu " .. single_kanji .. " karakterini içeren sözcükler")
				else
					table.insert(categories, "Japonca " .. single_kanji .. " karakterini içeren sözcükler")
				end
			else
				-- ]
				if yomi_type_by_kanji ~= 'düzensiz' and yomi_type_by_kanji ~= 'jukujikun' then
					track("no reading")
				end
				table.insert(categories, "Japonca " .. single_kanji .. " karakterini içeren sözcükler")
			end

			if reading_length <= 1 then
				table.insert(cell, "<small>" .. kanji_grade_links .. "</small>" .. ateji_text)
			else
				table.insert(cells_below, "| <small>" .. kanji_grade_links .. "</small>" .. ateji_text)
			end
		end
		table.insert(cells_above, table.concat(cell))
		kanji_pos = kanji_pos + reading_length
		if kanji_pos > kanji_length then break end
	end
	table.insert(cells, '|- style="background: white;"')
	if #cells_below > 0 then
		table.insert(cells, table.concat(cells_above, '\n'))
		table.insert(cells, '|- style="background: white;"')
		table.insert(cells, table.concat(cells_below, '\n'))
	else
		for i, v in ipairs(cells_above) do
			cells_above = v:gsub('| rowspan="2" | ', '| ')
		end
		table.insert(cells, table.concat(cells_above, '\n'))
	end

	local yomi_info = {
		 = {
			text = "on’yomi",
			entry = "Ek:Japonca_açıklamaları#on'yomi",
			category = "Japonca on'yomi ile okunan sözcükler",
		},
		 = {
			text = "kan’on",
			entry = "Ek:Japonca_açıklamaları#kan'on",
			category = "Japonca on'yomi ile okunan sözcükler",
		},
		 = {
			text = "goon",
			entry = "Ek:Japonca_açıklamaları#goon",
			category = "Japonca on'yomi ile okunan sözcükler",
		},
		 = {
			text = "tōon",
			entry = "Ek:Japonca_açıklamaları#tōon",
			category = "Japonca on'yomi ile okunan sözcükler",
		},
		 = {
			text = "kun’yomi",
			entry = "Ek:Japonca_açıklamaları#kun'yomi",
			category = "Japonca kun'yomi ile okunan sözcükler",
		},
		 = {
			text = "nanori",
			entry = "Ek:Japonca_açıklamaları#nanori",
			category = "Japonca nanori ile okunan sözcükler",
		},
		 = {
			text = "yutōyomi",
			entry = "Ek:Japonca_açıklamaları#yutōyomi",
			category = "Japonca yutōyomi ile okunan sözcükler",
		},
		 = {
			text = "jūbakoyomi",
			entry = "Ek:Japonca_açıklamaları#jūbakoyomi",
			category = "Japonca jūbakoyomi ile okunan sözcükler",
		},
		 = {
			text = "jukujikun",
			entry = "Ek:Japonca_açıklamaları#jukujikun",
			category = "Japonca jukujikun ile okunan sözcükler",
		},
		 = {
			text = "''düzensiz''",
			category = "Japonca düzensiz kanji okumalı sözcükler",
		},
		 = {
			text = "kan’yōon",
			entry = "Ek:Japonca_açıklamaları#kan'yoon",
			category = "Japonca kan'yōon ile okunan  sözcükler",
		},
	}

	local rendaku = args.r
	if rendaku then
		table.insert(categories, "Japonca rendaku ile okunan sözcükler")
	end

	if yomi then
		table.insert(cells, "|-")
		for _, i in ipairs(yomi) do
			local yomi_info = yomi_info] or { text = i }
			local text
			if yomi_info.entry then
				text = "]"
			else
				text = yomi_info.text
			end
			table.insert(cells, '| colspan="' .. i .. '" |' .. text)
		end
		local is_onyomi = { on = true, kanon = true, goon = true, toon = true, kanyoon = true }
		-- categories
		local all_onyomi = true
		for i = 1, #yomi do
			if not is_onyomi] then all_onyomi = false; break end
		end
		if all_onyomi then
			table.insert(categories, yomi_info.on.category)
		elseif yomi == 'jūbakoyomi' or yomi == 'yutōyomi' then
			table.insert(categories, yomi_info].category)
		else
			local all_yomi_of_same_type = true
			for i = 2, #yomi do
				if yomi ~= yomi then all_yomi_of_same_type = false; break end
			end
			if all_yomi_of_same_type then
				table.insert(categories, yomi_info].category)
			elseif #yomi == 2 and yomi == 1 and yomi == 1 and mw.ustring.len(pagename) == 2 then
				if is_onyomi] and yomi == 'kun' then
					table.insert(categories, yomi_info.category)
				elseif yomi == 'kun' and is_onyomi] then
					table.insert(categories, yomi_info.category)
				end
			end
		end
	end

	local kanji_table = kanji_length == 0 and "" or (table_head ..
		mw.ustring.gsub(kanji, '(.)', '| style="padding: 0.5em;" | ]\n') ..
		table.concat(cells, '\n') ..
		'\n|}')

	local forms_table = ""
	if args.alt and args.alt ~= "" and args.alt ~= "-" then
		local forms = {}
		for form in mw.text.gsplit(args.alt, ',') do
			local i_semicolon = string.find(form, ':')
			if i_semicolon then
				local altform = string.sub(form, 1, i_semicolon - 1)
				local altlabels = mw.text.split(string.sub(form, i_semicolon + 1), ' ')
				table.insert(forms, '<span class="Jpan" lang="ja" style="font-size:140%">]</span>')
			else
				table.insert(forms, '<span class="Jpan" lang="ja" style="font-size:140%">]</span>')
			end
		end
		forms_table = '\n' .. [[{| class="wikitable floatright"
! style="font-weight:normal" | Farklı yazılış]] .. (#forms == 1 and "" or "lar") .. '\n' .. [[
| style="text-align:center;font-size:108%" | ]] .. table.concat(forms, '<br>') .. '\n|}'
	end


	-- use user-provided sortkey if we got one, otherwise
	-- use the sortkey we've already made by combining the
	-- readings if provided, if we have neither then
	-- default to empty string and don't sort
	local userprovided_sortkey = args.sort
	if userprovided_sortkey then
		sortkey = userprovided_sortkey
	end
	if sortkey then
		sortkey = m_ja.jsort(sortkey)
	end

	return (forms_table == "" and kanji_table or (kanji_table .. forms_table)) .. m_utilities.format_categories(categories, lang, sortkey)
end

return export