Mô đun:Ruby

Có thể viết tài liệu về mô đun này tại Mô đun:Ruby/tài liệu.

--- Lexicographic tools for Vietnamese language text.

local m_str_utils = require("Module:string utilities")

local find   = m_str_utils.find
local format = string.format
local gmatch = m_str_utils.gmatch
local gsub   = m_str_utils.gsub
local len    = m_str_utils.len
local match  = m_str_utils.match
local upper  = m_str_utils.upper

local lang   = require("Module:languages").getByCode("vi")

local p = {}

---Converts the given text to traditional tone marks.
function p.toTraditionalTones(text)
	if type(text) == "table" then
		text = text.args[1]
	end
	return gsub(text, "%a+", function (word)
		if match(word, "^qu[yýỳỷỹỵ]$") then return word end
		return gsub(word, "%a%a$", {
			["oá"] = "óa", ["oà"] = "òa", ["oả"] = "ỏa", ["oã"] = "õa", ["oạ"] = "ọa",
			["oé"] = "óe", ["oè"] = "òe", ["oẻ"] = "ỏe", ["oẽ"] = "õe", ["oẹ"] = "ọe",
			["uý"] = "úy", ["uỳ"] = "ùy", ["uỷ"] = "ủy", ["uỹ"] = "ũy", ["uỵ"] = "ụy"
		})
	end)
end

---Converts the given text to reformed tone marks.
function p.toReformedTones(text)
	if type(text) == "table" then
		text = text.args[1]
	end
	return gsub(text, "%a+", function (word)
		return gsub(word, "%a%a$", {
			["óa"] = "oá", ["òa"] = "oà", ["ỏa"] = "oả", ["õa"] = "oã", ["ọa"] = "oạ",
			["óe"] = "oé", ["òe"] = "oè", ["ỏe"] = "oẻ", ["õe"] = "oẽ", ["ọe"] = "oẹ",
			["úy"] = "uý", ["ùy"] = "uỳ", ["ủy"] = "uỷ", ["ũy"] = "uỹ", ["ụy"] = "uỵ"
		})
	end)
end

---Generate alternative orthographies.
function p.allSpellings(main_spelling, makeLinks)
	local frame = nil
	if type(main_spelling) == "table" then
		frame = main_spelling
		main_spelling, makeLinks = frame.args[1], frame.args.link
	end

	local xformers = {
		p.toTraditionalTones,
		p.toReformedTones,
	}

	local spellings = {}
	for _, xformer in ipairs(xformers) do
		local alt_spelling = xformer(main_spelling)
		if not spellings[alt_spelling] then
			table.insert(spellings, alt_spelling)
			spellings[alt_spelling] = true
		end
	end

	if makeLinks then
		local m_links = require("Module:links")
		for k, link in ipairs(spellings) do
			spellings[k] = m_links.full_link({lang = lang, term = link})
		end
	end

	return frame and table.concat(spellings, "/") or spellings
end

---Unicode codepoints for combining Vietnamese tone marks.
p.combiningToneMarks = mw.ustring.char(0x300, 0x301, 0x303, 0x309, 0x323)

---Unicode codepoints for combining Vietnamese accent marks.
p.combiningAccentMarks = mw.ustring.char(0x302, 0x306, 0x31b)

---Strips Vietnamese diacritical marks from the given text.
function p.removeDiacritics(text, toneMarks, accentMarks, stroke)
	if type(text) == "table" then
		text, toneMarks, accentMarks, stroke = text.args[1],
			not text.args.tones or tonumber(text.args.tones) == 1,
			not text.args.accents or tonumber(text.args.accents) == 1,
			not text.args["đ"] or tonumber(text.args["đ"]) == 1
	end

	text = mw.ustring.toNFD(text)

	if toneMarks then
		text = gsub(text, "[" .. p.combiningToneMarks .. "]", "")
	end
	if accentMarks then
		text = gsub(text, "[" .. p.combiningAccentMarks .. "]", "")
	end
	if stroke then
		text = gsub(text, "[Đđ]", {["Đ"] = "D", ["đ"] = "d"})
	end

	return mw.ustring.toNFC(text)
end

---Vietnamese letters for use in comp().
p.letters = "aAàÀảẢãÃáÁạẠăĂằẰẳẲẵẴắẮặẶâÂầẦẩẨẫẪấẤậẬbBcCdDđĐeEèÈẻẺẽẼéÉẹẸêÊềỀểỂễỄếẾệỆfFgGhHiIìÌỉỈĩĨíÍịỊjJkKlLmMnNoOòÒỏỎõÕóÓọỌôÔồỒổỔỗỖốỐộỘơƠờỜởỞỡỠớỚợỢpPqQrRsStTuUùÙủỦũŨúÚụỤưƯừỪửỬữỮứỨựỰvVwWxXyYỳỲỷỶỹỸýÝỵỴzZ"

---Compare two syllables according to Vietnamese dictionary sorting order.
function p.compWord(word1, word2)
	if find(word1, word2, 1, true) == 1 then return false end
	if find(word2, word1, 1, true) == 1 then return true end

	local iter1 = gmatch(word1, "[" .. p.letters .. "]")
	local iter2 = gmatch(word2, "[" .. p.letters .. "]")

	while true do
		local c1 = iter1()
		local c2 = iter2()
		if c1 == nil or c2 == nil then break end

		local idx1 = find(p.letters, c1, 1, true)
		local idx2 = find(p.letters, c2, 1, true)

		if idx1 and idx2 then
			if idx1 < idx2 then return true end
			if idx1 > idx2 then return false end
		end
	end

	return word1 < word2
end

---Abbreviations and text for Han tu references.
p.refAbbreviations = {
	tvctdhv = "Trần (1999)",
	hvttd   = "Nguyễn (1974)",
	vntd    = "Văn Mới (1954)",
	tchvtd  = "Thiều Chửu (1942)",
	tdcndg  = "Nguyễn (2014)",
	tdcntd  = "Nguyễn et al. (2009)",
	gdhn    = "Trần (2004)",
	dtdcn   = "Vũ (1998)",
	btcn    = "Hồ (1976)",
	bonet   = "Bonet (1899)",
	genibrel= "Génibrel (1898)",
	taberd  = "Taberd & Pigneau de Béhaine (1838)",
}

---Creates a ref tag containing [[Template:vi-ref]].
function p.createRefTag(ref)
	local refFullName = p.refAbbreviations[ref] or ref
	return mw.getCurrentFrame():extensionTag{
		name = "ref",
		args = { name = ref },
		content = format("{{vi-ref|%s.}}", refFullName)
	}
end

---[[Template:vi-readings]]
function p.readings(hanviet, nom, rs, phienthiet, reading)
	local pagename = mw.loadData("Module:headword/data").pagename

	if type(hanviet) == "table" then
		local args = hanviet:getParent().args
		hanviet, nom, rs, phienthiet, reading =
			args.hanviet or args.hv, args.nom or args.n, args.rs or args.sort,
			args.phienthiet or args.phth or args.fanqie, args.reading or args.readings
	end

	local lines = {}
	local styles = {
		{
			link = "Hán Việt",
			cat = "Vietnamese Chữ Hán",
			list = hanviet and mw.text.split(hanviet, "%s*,%s*"),
			phienthiet = phienthiet and mw.text.split(phienthiet, "%s*,%s*")
		},
		{
			link = "chữ Nôm|Nôm",
			cat = "Vietnamese Nom",
			list = nom and mw.text.split(nom, "%s*,%s*"),
		},
		{
			link = "Hán Nôm",
			cat = "Vietnamese Han characters with unconfirmed readings",
			list = reading and mw.text.split(reading, "%s*,%s*")
		},
	}

	for _, style in ipairs(styles) do
		if style.list and #style.list > 0 and #style.list[1] > 0 then
			local readings = style.list

			for j, reading in ipairs(readings) do
				local ref
				local a, b = match(reading, "(.-)%s*%-%s*(.+)")
				if a then
					reading, ref = a, b
				end

				local spellings = p.allSpellings(reading, true)
				readings[j] = table.concat(spellings, "/")

				-- Linking of "切" to "fanqie"
				if style.phienthiet and style.phienthiet[j] then
					local pruby = "link"
					local ruby = p.ruby(match(mw.text.trim(style.phienthiet[j]), "(%a+) +(.+)"))
					if ruby then
						pruby = "nocolor"
						local suffix = p.ruby("切", "thiết")
						readings[j] = format("%s (%s[[fanqie#English|%s]])", readings[j], ruby, suffix)
					end
				end

				-- References
				if ref then
					for r in mw.text.gsplit(ref, "%s*;%s*") do
						readings[j] = readings[j] .. p.createRefTag(r)
					end
				end
			end

			if #readings > 0 then
				local sortkey = rs or pagename
				readings = table.concat(readings, ", ")
				table.insert(lines, format(
					"<span class='Hani' lang='vi' style='font-size: 135%%;'>%s</span>: '''[[%s]]''' readings: %s[[Category:%s|%s]] [[Category:Vietnamese lemmas]] [[Category:Vietnamese Han characters]]</br>",
					pagename, style.link, readings, style.cat, sortkey))
			end
		end
	end

	return table.concat(lines, "\n")
end

---[[Template:vi-ruby]]
local pruby = {}  -- shared state used by readings() and ruby()

function p.ruby(characters, readings, mark, alts)
	if type(characters) == "table" then
		local args = characters:getParent().args
		characters, readings, mark, alts =
			args[1] or "",
			args[2] or "",
			args.mark or mw.loadData("Module:headword/data").pagename,
			((args.alts and mw.text.split(args.alts, "%s+")) or
			 (args.ids and mw.text.split(args.ids, "%s+")) or {})
	end

	if not readings then
		return characters
	end

	readings = mw.text.split(readings, "[^" .. p.letters .. "]+")
	local result = {}
	local character_idx = 1
	local alt_idx = 1

	for character in gmatch(characters, ".") do
		local is_alt = false
		if character == "*" and alts[alt_idx] then
			character = alts[alt_idx]
			is_alt = true
			alt_idx = alt_idx + 1
		end

		if is_alt or (match(character, "^%a$") and not character:match("^%w$")) then
			local reading = readings[character_idx]

			if mark and character == mark then
				character = format("<mark>%s</mark>", character)
				reading = format("<mark>%s</mark>", reading)
			end

			if pruby == 'link' then
				character = format(
					"<ruby><rb><span class='Hani' style='font-size: 100%%'>[[%s#Vietnamese|%s]]</span></rb><rp>(</rp><rt><span style='padding: 0 0.25em; font-size: 135%%;'>[[%s#Vietnamese|%s]]</span></rt><rp>)</rp></ruby>",
					character, character, reading, reading)
			elseif pruby == 'nocolor' then
				character = format(
					"<ruby><rb><span class='Hani' style='color:#202122;'>%s</span></rb><rp>(</rp><rt><span style='padding: 0 0.25em; font-size: 125%%;'>%s</span></rt><rp>)</rp></ruby>",
					character, reading)
			else
				character = format(
					"<ruby><rb><span class='Hani'>%s</span></rb><rp>(</rp><rt><span style='padding: 0 0.25em;'>%s</span></rt><rp>)</rp></ruby>",
					character, reading)
			end

			character_idx = character_idx + 1
		end
		table.insert(result, character)
	end

	return format("<span lang='vi' style='font-size: 135%%;'>%s</span>", table.concat(result))
end

function p.hantutab()
	local hantu = gsub(mw.loadData("Module:headword/data").pagename, '[^一-鿿㐀-䶿﨎﨏﨑﨓﨔﨟﨡﨣﨤﨧-﨩𠀀-𪛟𪜀-𮯯𰀀-𱍏]', '')
	local table_head = '<table class="floatright wikitable" style="text-align:center; font-size:small;"><tr><th colspan="' ..
		len(hantu) ..
		'" style="font-weight:normal;">[[Hán tự]] in this term</th></tr><tr lang="vi" class="Hani" style="font-size:2em; background:white; line-height:1em;">'

	return table_head ..
		gsub(hantu, '(.)', '<td style="padding:0.5em;">[[%1#Vietnamese|%1]]</td>') ..
		'</tr></table>'
end

---Returns the categories indicated by the given wikitext.
function p.classifierCategories(frame)
	local src = frame.args[1]
	local classifiers = {}
	for classifier in gmatch(gsub(src, "<[^>]->", ""), "[" .. p.letters .. "]+") do
		if classifier ~= "l" and classifier ~= "vi" and classifier ~= "vi-l" and
		   classifier ~= "Vietnamese" then
			local cat = format("[[Category:Vietnamese nouns classified by %s]]", classifier)
			table.insert(classifiers, cat)
		end
	end
	return table.concat(classifiers)
end

function p.new(frame)
	local title = mw.loadData("Module:headword/data").pagename
	local args = frame:getParent().args
	local pos = args[1] or ""
	local def = args[2] or "{{rfdef|vi}}"
	local pos2 = args[3] or (args[4] and "" or false)
	local def2 = args[4] or "{{rfdef|vi}}"
	local pos3 = args[5] or (args[6] and "" or false)
	local def3 = args[6] or "{{rfdef|vi}}"
	local etym = args["e"] or false
	local head = args["head"] or false
	local cat = args["cat"] or false
	local cls = args["cls"] or false
	local rdp = args["rdp"] or false
	local nom = args["nom"] or false
	local pic = args["pic"] or false
	local picc = args["picc"] or false

	nom = nom and gsub(nom, "(.)", "[[%1]], ") or false
	nom = nom and gsub(nom, ", $", "") or false

	if args["h"] then
		etym = "{{vi-etym-sino|" .. args["h"] .. "}}."
	end

	if not etym and match(title, " ") then
		etym = "{{com|vi"
		for word in mw.text.gsplit(title, " ") do
			etym = etym .. "|" .. word
		end
		etym = etym .. "}}."
	end

	if etym == "-" then etym = false end
	if etym then etym = gsub(etym, "^%<", "From") end

	local result = ""

	local function genTitle(text)
		local pos_title = {
			[""] = "Noun", ["n"] = "Noun", ["pn"] = "Proper noun", ["propn"] = "Proper noun", ["pron"] = "Pronoun",
			["v"] = "Verb", ["vf"] = "Verb", ["a"] = "Adjective", ["adj"] = "Adjective", ["adv"] = "Adverb",
			["prep"] = "Preposition", ["postp"] = "Postposition", ["conj"] = "Conjunction",
			["part"] = "Particle", ["suf"] = "Suffix",
			["prov"] = "Proverb", ["id"] = "Idiom", ["ph"] = "Phrase", ["intj"] = "Interjection",
			["cl"] = "Classifier", ["cls"] = "Classifier", ["num"] = "Numeral", ["abb"] = "Abbreviation", ["deter"] = "Determiner"
		}
		return pos_title[text] or upper(text:sub(1,1)) .. text:sub(2)
	end

	local function genHead(text)
		local pos_head = {
			[""] = "noun", ["n"] = "noun", ["pn"] = "proper noun", ["propn"] = "proper noun",
			["v"] = "verb", ["vf"] = "verb form", ["a"] = "adj",
			["postp"] = "post", ["conj"] = "conj", ["part"] = "particle", ["pron"] = "pronoun",
			["prov"] = "proverb", ["id"] = "idiom", ["ph"] = "phrase", ["intj"] = "interj",
			["abb"] = "abbr", ["cl"] = "classifier", ["deter"] = "det"
		}
		return pos_head[text] or text
	end

	local function other(class, title, args)
		local code = ""
		local i = 2
		if class == "der" and args[class] then
			code = code .. "\n\n===" .. title .. "===\n{{col3|vi|" .. args[class]
			while args[class .. i] do
				code = code .. "|" .. args[class .. i]
				i = i + 1
			end
			code = code .. "}}"
		elseif args[class] then
			code = code .. "\n\n===" .. title .. "===\n* {{l|vi|" .. args[class] .. "}}"
			while args[class .. i] do
				code = code .. "\n* {{l|vi|" .. args[class .. i] .. "}}"
				i = i + 1
			end
		end
		return code
	end

	result = result .. "==Vietnamese=="
	if args["wp"] then
		result = result .. "\n{{wp|vi:" .. (args["wp"] == "y" and "" or args["wp"]) .. "}}"
	end
	if pic then
		result = result .. "\n[[File:" .. pic .. "|thumb|" .. (picc or upper(title:sub(1,1)) .. title:sub(2) .. ".") .. "]]"
	end

	result = result .. other("alt", "Alternative forms", args)
	if etym then result = result .. "\n\n===Etymology===\n" .. etym end
	result = result .. "\n\n===Pronunciation===\n{{vi-IPA}}"
	result = result .. "\n\n===" .. genTitle(pos) .. "===\n{{vi-" .. genHead(pos) ..
		(head and ("|head=" .. head) or "") ..
		((genHead(pos) == "noun" and cls) and "|cls=" .. cls or "") ..
		(((genHead(pos) == "adj" or genHead(pos) == "verb" or genHead(pos) == "adv") and rdp) and "|rdp=" .. rdp or "") ..
		(nom and "|" .. nom or "") ..
		"}}\n\n# " .. def

	result = result .. other("syn", "=Synonyms=", args)
	result = result .. other("ant", "=Antonyms=", args)
	result = result .. other("der", "=Derived terms=", args)
	result = result .. other("also", "=See also=", args)

	if pos2 then
		result = result .. "\n\n===" .. genTitle(pos2) .. "===\n{{vi-" .. genHead(pos2) ..
			(head and ("|head=" .. head) or "") ..
			((genHead(pos) == "noun" and cls) and "|cls=" .. cls or "") ..
			(((genHead(pos2) == "adj" or genHead(pos2) == "verb" or genHead(pos2) == "adv") and rdp) and "|rdp=" .. rdp or "") ..
			(nom and "|" .. nom or "") ..
			"}}\n\n# " .. def2
	end

	if pos3 then
		result = result .. "\n\n===" .. genTitle(pos3) .. "===\n{{vi-" .. genHead(pos3) ..
			(head and ("|head=" .. head) or "") ..
			((genHead(pos) == "noun" and cls) and "|cls=" .. cls or "") ..
			(((genHead(pos3) == "adj" or genHead(pos3) == "verb" or genHead(pos3) == "adv") and rdp) and "|rdp=" .. rdp or "") ..
			(nom and "|" .. nom or "") ..
			"}}\n\n# " .. def3
	end

	if cat then result = result .. "\n\n{{C|vi|" .. cat .. "}}" end

	return result
end

function p.new_der(frame)
	local title = mw.loadData("Module:headword/data").pagename
	local data_module = require("Module:vi/vocab-list")
	local args = frame:getParent().args
	local result = {}

	for _, arg in ipairs(args) do
		table.insert(result, arg)
	end

	for _, word in ipairs(data_module) do
		if find(word, title) and word ~= title and not find(word, title .. "[^ ]") and not find(word, "[^ ]" .. title) then
			table.insert(result, word)
		end
	end

	local hash, res = {}, {}
	for _, element in ipairs(result) do
		if not hash[element] then
			res[#res + 1] = element
			hash[element] = true
		end
	end

	local vi_sort_module = require("Module:vi-sortkey")
	local makeSortKey = require("Module:memoize")(vi_sort_module.makeSortKey)

	table.sort(res, function(term1, term2)
		return makeSortKey(term1) < makeSortKey(term2)
	end)

	return "{{vi-der|" .. table.concat(res, "|") .. "}}"
end

function p.derived(frame)
	local tu_lay_note = "<span style=\"padding-left:4px; padding-right:4px\">&nbsp;</span><span style=\"background:#ffffe0\">(''[[từ láy]]'')</span>"
	local m_columns = require("Module:columns")
	local m_links = require("Module:links")
	local args = frame:getParent().args
	local pagename = mw.loadData("Module:headword/data").pagename
	local result = {}
	local length = 0

	local unfold = args["unfold"] and true or false
	local title = args["title"] or false
	local title_text = title or "Derived terms"

	for _, word in ipairs(args) do
		local w, is_tu_lay = gsub(word, "%:tl", "")
		local tu_lay = is_tu_lay > 0 and tu_lay_note or ""
		local word_parts = mw.text.split(gsub(w, "\n", ""), ":")
		table.insert(result, m_links.full_link({
			lang = lang,
			term = word_parts[1],
			gloss = word_parts[2] or nil
		}) .. tu_lay)
		length = math.max(len(w), length)
	end

	return m_columns.create_list {
		column_count = (length > 15 and 2 or 3),
		content = result,
		header = "''(" .. title_text .. ")'':",
		alphabetize = true,
		collapse = (unfold or #result < 7) and false or true,
		class = "columns-bg",
		lang = lang
	}
end

return p