Module:kanjitab

Ang dokumentasyon ng ng modyul na ito ay maaaring likhain sa Module:kanjitab/doc

local export = {}

local m_str_utils = require("Module:string utilities")
local m_utilities = require("Module:utilities")
local m_ja = require("Module:ja")
local show_labels = require("Module:labels").show_labels
--[=[
	Other modules used: [[Module:parameters]]
]=]

local concat = table.concat
local convert_iteration_marks = require("Module:Hani").convert_iteration_marks
local find = string.find
local gsplit = m_str_utils.gsplit
local gsub = string.gsub
local kata_to_hira = m_ja.kata_to_hira
local insert = table.insert
local match = string.match
local remove = table.remove
local split = m_str_utils.split
local sub = string.sub
local ugsub = mw.ustring.gsub
local ulen = m_str_utils.len
local umatch = mw.ustring.match
local usub = m_str_utils.sub

local PAGENAME = mw.loadData("Module:headword/data").pagename
local NAMESPACE = mw.title.getCurrentTitle().nsText

local d_range = mw.loadData("Module:ja/data/range")
local yomi_data = mw.loadData("Module:kanjitab/data")

local kanji_grade_links = {
	"[[Appendix:Japanese_glossary#kyōiku_kanji|Grade: 1]]",
	"[[Appendix:Japanese_glossary#kyōiku_kanji|Grade: 2]]",
	"[[Appendix:Japanese_glossary#kyōiku_kanji|Grade: 3]]",
	"[[Appendix:Japanese_glossary#kyōiku_kanji|Grade: 4]]",
	"[[Appendix:Japanese_glossary#kyōiku_kanji|Grade: 5]]",
	"[[Appendix:Japanese_glossary#kyōiku_kanji|Grade: 6]]",
	"[[Appendix:Japanese_glossary#jōyō_kanji|Grade: S]]",		-- 7
	"[[Appendix:Japanese_glossary#jinmeiyō_kanji|Jinmeiyō]]",	-- 8
	"[[Appendix:Japanese_glossary#hyōgai_kanji|Hyōgai]]"		-- 9
}

-- this is the function that is called from templates
function export.show(frame)
	local args = require("Module:parameters").process(frame:getParent().args, {
		[1] = { list = true, allow_holes = true },
		k = { list = true, allow_holes = true },
		o = { list = true, allow_holes = true },
		r = {},
		sort = {},
		yomi = {},
		ateji = {},
		alt = {},
		alt2 = {},
		kyu = { list = true },
		y = {alias_of = "yomi"},
		clearright = {type = "boolean"},
		pagename = {},
	})
	
	local lang_code = frame.args[1]
	local lang = require("Module:languages").getByCode(lang_code)
	local lang_name = lang:getCanonicalName()

	if args.pagename and NAMESPACE == "" then
		require("Module:debug/track")("kanjitab/pagename param in mainspace")
	end
	local pagename = args.pagename or PAGENAME

	local categories = {}
	local cells = {}

	-- extract kanji and non-kanji
	local kanji = {}
	local non_kanji = {}
	
	-- 々 and 〻
	pagename = convert_iteration_marks(pagename)
	
	local kanji_border = 1
	ugsub(pagename, "()([" .. d_range.kanji .. "々〻])()", function(p1, w1, p2)
		insert(non_kanji, usub(pagename, kanji_border, p1 - 1))
		kanji_border = p2
		insert(kanji, w1)
	end)
	insert(non_kanji, usub(pagename, kanji_border))

	-- kyujitai
	local kyu = args.kyu
	if kyu[1] == "-" then
		kyu = {}
	elseif kyu[1] == nil then
		local form_kyu = {non_kanji[1]}
		local kyu_data = mw.loadData("Module:ja/data/kyu")
		local has_kyu, has_kyu_nonsupple, has_shin = false, false, false
		for i, v in ipairs(kanji) do
			local v_kyu = match(kyu_data[1], v .. "(%S*)%s")
			if v_kyu == nil then
				insert(form_kyu, v)
			elseif v_kyu == "" then
				has_shin = true
				break
			elseif v_kyu:sub(1, 1) == "&" then
				has_kyu = true
				insert(form_kyu, v_kyu)
			else
				has_kyu, has_kyu_nonsupple = true, true
				insert(form_kyu, v_kyu)
			end
			insert(form_kyu, non_kanji[i + 1])
		end

		if not has_shin and has_kyu then
			kyu[1] = (has_kyu_nonsupple and "" or pagename .. "|") .. concat(form_kyu)
		end

		if find(pagename, "弁") then
			require("Module:debug/track")("kanjitab/ambiguous kyujitai for 弁")
			kyu[1] = "which 弁?"
		end
	end
	
	local all_yomi, missing_yomi
	if args.yomi then
		all_yomi = {}
		
		local keys = split(args.yomi, ",")
		for i, yomi, len in ipairs(keys) do
			yomi, len = match(yomi, "^(%l*)(%d*)$")
			yomi = yomi_data[yomi] or error("The yomi type \"" .. yomi .. "\" in the input \"" .. args.yomi .. "\" is not recognized.")
			if len ~= "" then
				-- Disallow length 0 or leading zeroes, as a sanity check.
				len = match(len, "^[1-9]%d*$") and tonumber(len) or error("Cannot specify a length of " .. len .. " kanji.")
			-- Only one yomi with no length given: apply to all kanji.
			elseif i == 1 and #keys == 1 then
				len = #kanji
			else
				len = 1
			end
			
			local yomi_type = yomi.type
			-- If the on'yomi is not specified as goon/kanon/toon/soon, only "on".
			if yomi_type == "on'yomi" then
				require("Module:debug/track")("kanjitab/unspecified on")
			elseif yomi_type == "jūbakoyomi" then
				require("Module:debug/track")("kanjitab/jubakoyomi")
			elseif yomi_type == "yutōyomi" then
				require("Module:debug/track")("kanjitab/yutoyomi")
			end
			
			-- If the yomi requires a specific number of kanji (e.g. jūbakoyomi, yutōyomi).
			local req_kanji = yomi.required_kanji
			if req_kanji and #kanji ~= req_kanji then
				error("The yomi type \"" .. yomi.type .. "\" is only applicable to terms with " .. req_kanji .. " kanji.")
			elseif yomi.type == "none" then
				missing_yomi = true
			end
			
			-- Insert yomi data for each applicable kanji. Wrap in a table first, as the range for this input yomi is determined by its identity, so that (e.g.) "kun,kun" is still treated as two separate inputs.
			yomi = {data = yomi}
			for _ = 1, len do
				insert(all_yomi, yomi)
			end
		end
		-- If there are any yomi slots left, handle them as empty.
		if #all_yomi < #kanji then
			missing_yomi = true
			for _ = #all_yomi + 1, #kanji do
				insert(all_yomi, {data = yomi_data.none})
			end
		end
	elseif #kanji > 0 then
		missing_yomi = true
	end
	
	if missing_yomi then
		insert(categories, lang_name .. " terms with missing yomi")
	end
	
	-- process readings
	local readings = {}
	local readings_actual = {}
	local reading_length_total = 0
	for i = 1, args[1].maxindex do
		local reading_kana, reading_length = match(args[1][i] or "", "^(%D*)(%d*)$")
		reading_kana = reading_kana ~= "" and reading_kana or nil
		reading_length = reading_kana and tonumber(reading_length) or 1

		insert(readings, {reading_kana, reading_length})
		reading_length_total = reading_length_total + reading_length
	end
	if reading_length_total > #kanji then
		error("Readings for " .. reading_length_total .. " kanji are given, but this word has only " .. #kanji .. " kanji.")
	else
		for _ = reading_length_total + 1, #kanji do
			insert(readings, {nil, 1})
		end
	end

	local table_head = [=[
{| class="wikitable kanji-table floatright" style="text-align: center; ]=] .. (args.clearright and " clear:right;" or "") .. [=["
! ]=] .. (#kanji > 1 and "colspan=\"" .. #kanji .. "\" " or "") .. [=[style="font-weight: normal;" | [[Appendix:Japanese_glossary#kanji|Kanji]] in this term
|- lang="]=] .. lang_code .. [=[" class="Jpan" style="font-size: 2em; background: white; line-height: 1em;"

]=]
	
	if args.k.maxindex and args.k.maxindex > args[1].maxindex then
		error("kanjitab/too many k")
	end
	
	if args.o.maxindex and args.o.maxindex > args[1].maxindex then
		error("kanjitab/too many o")
	end
	
	local is_ateji = {}
	if args.ateji then
		local ateji = args.ateji
		local cat_ateji = false
		if ateji == "y" then
			for i = 1, #kanji do
				is_ateji[i] = true
			end
			cat_ateji = true
		else
			for i in gsplit(ateji, ";") do
				gsub(i, "^(%d+)$", function(a)
					is_ateji[tonumber(a)] = true
					cat_ateji = true
				end)
				gsub(i, "^(%d+),(%d+)$", function (a, b)
					for j = tonumber(a), tonumber(b) do
						is_ateji[j] = true
					end
					cat_ateji = true
				end)
			end
		end
		if cat_ateji then insert(categories, lang_name .. " terms spelled with ateji") 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, reading in ipairs(readings) do
		local reading_kana, reading_length = reading[1], reading[2]
		local cell = {}

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

		-- display reading, actual reading and okurigana
		if reading_kana then
			if reading_kana ~= "" and reading_kana ~= "-" and umatch(reading_kana, "[^" .. d_range.kana .. "]") then
				error("Please remove any non-kana characters from the reading input " .. reading_kana .. ".")
			end

			local actual_reading = args.k[i]
			local okurigana = args.o[i]

			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

			readings_actual[i] = {(actual_reading or reading_kana) .. (okurigana or ""), reading_length}

			insert(cell, "<span class=\"Jpan\" lang=\"" .. lang_code .. "\">" .. text .. "</span>")
			if reading_length <= 1 then insert(cell, "<br/>") end
		else
			readings_actual[i] = {nil, 1}
		end

		-- display kanji grade, categorize
		for j = kanji_pos, kanji_pos + reading_length - 1 do
			local single_kanji = kanji[j]
			local kanji_grade = m_ja.kanji_grade(single_kanji)
			local ateji_text = is_ateji[j] and "<br/><small>([[Appendix:Japanese glossary#ateji|ateji]])</small>" or ""
			
			local type, compound
			if all_yomi then
				local yomi = all_yomi[j].data
				type, compound = yomi.type, yomi.compound_reading
			end
			
			if not reading_kana then
				if type ~= "irregular" then
					require("Module:debug/track")("kanjitab/no reading")
				end
				insert(categories, lang_name .. " terms spelled with " .. single_kanji)
			elseif reading_length ~= 1 or type == "irregular" then
				insert(categories, lang_name .. " terms spelled with " .. single_kanji)
			elseif compound then
				-- Re-enable once all bad jukujikun calls are fixed.
				-- error("The yomi type \"" .. type .. "\" is only applicable to compound character readings, so cannot apply to " .. single_kanji .. " read as " .. reading_kana .. ". If this is intended as part of a " .. type .. " reading, please enter the whole reading as one, followed by the number of kanji it applies to.")
				require("Module:debug/track")("kanjitab/single kanji with jukujikun")
			else -- Subcategorize by reading.
				insert(categories, lang_name .. " terms spelled with " .. single_kanji .. " read as " .. kata_to_hira(reading_kana))
			end

			if reading_length <= 1 then
				insert(cell, "<small>" .. kanji_grade_links[kanji_grade] .. "</small>" .. ateji_text)
			else
				insert(cells_below, "| <small>" .. kanji_grade_links[kanji_grade] .. "</small>" .. ateji_text)
			end
		end
		insert(cells_above, concat(cell))
		kanji_pos = kanji_pos + reading_length
	end
	insert(cells, "|- style=\"background: white;\"")
	if #cells_below > 0 then
		insert(cells, concat(cells_above, "\n"))
		insert(cells, "|- style=\"background: white;\"")
		insert(cells, concat(cells_below, "\n"))
	else
		for i, v in ipairs(cells_above) do
			cells_above[i] = gsub(v, "| rowspan=\"2\" | ", "| ")
		end
		insert(cells, concat(cells_above, "\n"))
	end
	
	local rendaku = args.r
	if rendaku then
		insert(categories, lang_name .. " terms with rendaku")
	end
	
	if all_yomi then
		insert(cells, "|-")
		local len, all_on, yomi_cat = 1, true
		for i, yomi in ipairs(all_yomi) do
			-- If the next kanji has the same yomi table, it's part of the same range.
			if yomi == all_yomi[i + 1] then
				len = len + 1
			else
				yomi = yomi.data
				local yomi_type = yomi.type
				local display = yomi.display or yomi_type
				local appendix = yomi.appendix
				insert(cells, "| colspan=\"" .. len .. "\" |" .. (
					appendix == false and display or
					"[[Appendix:Japanese_glossary#" .. (appendix or yomi_type) .. "|" .. display .. "]]"
				))
				-- Categorise as irregular if any irregular yomi are found; otherwise, categorise if all yomi are of the same type. If yomi are of different types but are all on, on'yomi is used as a fallback.
				if yomi_cat ~= "irregular" then
					local cat_type = yomi_type
					if cat_type == "irregular" or yomi_cat == nil then
						yomi_cat = cat_type
					elseif yomi_cat ~= cat_type then
						yomi_cat = false
					end
					if not yomi.onyomi then
						all_on = false
					end
				end
				len = 1
			end
		end
		if yomi_cat then
			-- Check yomi_data first, in case cat_type is "irregular"; if no match, must be some other type, so get it from the first yomi in all_yomi, since not all yomi types are yomi_data keys.
			yomi_cat = yomi_data[yomi_cat] or all_yomi[1].data
		elseif all_on then
			yomi_cat = yomi_data.on
		elseif #all_yomi == 2 then
			local y1, y2 = all_yomi[1].data, all_yomi[2].data
			if ulen(pagename) == 2 then
				if y1.onyomi and y2.type == "kun'yomi" then
					yomi_cat = yomi_data.j -- jūbakoyomi
				elseif y1.type == "kun'yomi" and y2.onyomi then
					yomi_cat = yomi_data.y -- yutōyomi
				end
			end
		end
		if yomi_cat then
			local category = yomi_cat.reading_category
			if category ~= false then
				insert(categories, lang_name .. " " .. "terms read with " .. (category or yomi_cat.type))
			end
		end
	end

	local kanji_table
	if #kanji > 0 then
		kanji_table = table_head
		for _, v in ipairs(kanji) do
			kanji_table = kanji_table .. "| style=\"padding: 0.5em;\" | [[" .. v .. "#" .. lang_name .. "|" .. v .. "]]\n"
		end
		kanji_table = kanji_table .. concat(cells, "\n") .. "\n|}"
	else
		kanji_table = ""
	end

	local forms_table = ""
	if args.alt == "" or args.alt == "-" then args.alt = nil end
	if kyu[1] or args.alt then
		local forms = {}

		-- |kyu=
		if kyu[1] == "which 弁?" then
			insert(forms, "<strong class=\"error\" style=\"font-size:75%;\">Please specify the correct kyujitai for 弁 with the parameter \"kyu\".</strong>[[Category:Requests for cleanup in " .. lang_name .. " entries]]")
			remove(kyu, 1)
		end

		for _, form in ipairs(kyu) do
			local form_linkto, form_display = match(form, "^(.+)|(.+)$")
			if not form_linkto then form_linkto, form_display = form, form end
			insert(forms, concat{
				"<span class=\"Jpan\" lang=\"" .. lang_code .. "\" style=\"font-family:游ゴシック, HanaMinA, sans-serif; font-size:140%;\">[[",
				form_linkto,
				form_linkto == pagename and "|" or "#" .. lang_name .. "|",
				form_display,
				"]]</span> <small>",
				show_labels {labels = {"kyūjitai"}, lang = lang, nocat = true },
				"</small>",
			})
		end

		-- |alt=
		if args.alt then
			for form in gsplit(args.alt, ",") do
				local i_semicolon = find(form, ":")
				if i_semicolon then
					local altform = sub(form, 1, i_semicolon - 1)
					local altlabels = split(sub(form, i_semicolon + 1), " ")
					insert(forms, concat{
						"<span class=\"Jpan\" lang=\"" .. lang_code .. "\" style=\"font-size:140%\">[[",
						altform,
						"#" .. lang_name .. "|",
						altform,
						"]]</span> <small>",
						show_labels { labels = altlabels, lang = lang, nocat = true },
						"</small>",
					})
				else
					insert(forms, concat{
						"<span class=\"Jpan\" lang=\"" .. lang_code .. "\" style=\"font-size:140%\">[[",
						form,
						"#" .. lang_name .. "|",
						form,
						"]]</span>"
					})
				end
			end
		end

		forms_table = "\n" .. [[{| class="wikitable floatright"
! style="font-weight:normal" | Alternative spelling]] .. (#forms == 1 and "" or "s") .. [[

|-
| style="text-align:center;font-size:108%" | ]] .. concat(forms, "<br>") .. "\n|}"
	end
	
	local forms_table2 = ""
	if args.alt2 and args.alt2 ~= "" and args.alt2 ~= "-" then
		local forms2 = {}
		for form in gsplit(args.alt2, ",") do
			insert(forms2, "<span class=\"Jpan\" lang=\"" .. lang_code .. "\">[[" .. form .. "#" .. lang_name .. "|" .. form .. "]]</span>")
		end
		forms_table2 = "\n" .. [[{| class="wikitable floatright"
! style="font-weight:normal" | Variant form]] .. (#forms2 == 1 and "" or "s") .. "\n" .. [[
| style="text-align:center;font-size:140%" | ]] .. concat(forms2, "<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 sortkey
	if args.sort then
		sortkey = args.sort
	else
		sortkey = {non_kanji[1]}
		local id = 1
		for _, v in ipairs(readings_actual) do
			id = id + v[2]
			insert(sortkey, (v[1] or "") .. (non_kanji[id] or ""))
		end
		sortkey = concat(sortkey)
	end
	if sortkey == "" then
		sortkey = nil
	else
		sortkey = lang:makeSortKey(sortkey)
	end
	
	if sortkey ~= lang:makeSortKey(PAGENAME) then
		require("Module:debug/track"){"kanjitab/nonstandard sortkey", "kanjitab/nonstandard sortkey/" .. lang_code}
	end

	return kanji_table .. forms_table .. forms_table2 .. m_utilities.format_categories(categories, lang, sortkey)
end

return export