ข้ามไปเนื้อหา

มอดูล:Brah-translit

จาก วิกิพจนานุกรม พจนานุกรมเสรี

This module will transliterate text in the อักษรพราหมี. It is used to transliterate ปรากฤตแบบอโศก, Kamarupi Prakrit, โคตาน, ทมิฬเก่า, บาลี, ปรากฤต, สันสกฤต, และตุมซุก. The module should preferably not be called directly from templates or other modules. To use it from a template, use {{xlit}}. Within a module, use Module:languages#Language:transliterate.

For testcases, see Module:Brah-translit/testcases.

Functions

tr(text, lang, sc)
Transliterates a given piece of text written in the script specified by the code sc, and language specified by the code lang.
When the transliteration fails, returns nil.

local export = {}
local gsub = mw.ustring.gsub
local u = mw.ustring.char
local letter_with_mark = "(.["..u(0x0300).."-"..u(0x036F).."]?)"

local tt = {
	-- consonants
	["𑀓"] = "ก", ["𑀔"] = "ข", ["𑀕"] = "ค", ["𑀖"] = "ฆ", ["𑀗"] = "ง",
	["𑀘"] = "จ", ["𑀙"] = "ฉ", ["𑀚"] = "ช", ["𑀛"] = "ฌ", ["𑀜"] = "ญ",
	["𑀝"] = "ฏ", ["𑀞"] = "ฐ", ["𑀟"] = "ฑ", ["𑀠"] = "ฒ", ["𑀡"] = "ณ",
	["𑀢"] = "ต", ["𑀣"] = "ถ", ["𑀤"] = "ท", ["𑀥"] = "ธ", ["𑀦"] = "น",
	["𑀧"] = "ป", ["𑀨"] = "ผ", ["𑀩"] = "พ", ["𑀪"] = "ภ", ["𑀫"] = "ม",
	["𑀬"] = "ย", ["𑀭"] = "ร", ["𑀮"] = "ล", ["𑀯"] = "ว",
	["𑀰"] = "ศ", ["𑀱"] = "ษ", ["𑀲"] = "ส", ["𑀳"] = "ห", ["𑀴"] = "ฬ",
	["𑀵"] = "ฬ̱", ["𑀶"] = "ร̱", ["𑀷"] = "น̱",
	-- independent vowels
	["𑀅"] = "อ", ["𑀆"] = "อา", ["𑀇"] = "อิ", ["𑀈"] = "อี", ["𑀉"] = "อุ", ["𑀊"] = "อู",
	["𑀋"] = "ฤ", ["𑀌"] = "ฤๅ", ["𑀍"] = "ฦ", ["𑀎"] = "ฦๅ",
	["𑀏"] = "เอ", ["𑀐"] = "ไอ", ["𑀑"] = "โอ", ["𑀒"] = "เอา",
	-- dependent vowels and diacritics (excluding front type)
	["𑀸"] = "า", ["𑀹"] = "า", ["𑀺"] = "ิ", ["𑀻"] = "ี", ["𑀼"] = "ุ", ["𑀽"] = "ู",
	["𑀾"] = "ฺฤ", ["𑀿"] = "ฺฤๅ", ["𑁀"] = "ฺฦ", ["𑁁"] = "ฺฦๅ",
	["𑀁"] = "ํ", ["𑀂"] = "ห์", ["𑁆"] = "ฺ",
	["𑀀"] = "ม̐", -- candrabindu
	-- marks
	["𑀃"] = "⧖", ["𑀄"] = "⨂",
	["𑁇"] = "ฯ", ["𑁈"] = "๚", ["𑁉"] = "‧", ["𑁊"] = "⁚",
	["𑁋"] = "—", ["𑁌"] = "๛", ["𑁍"] = "๏",
	-- zero-width space (display it if it hides in a word)
	[u(0x200B)] = "‼",
}

local adjust1 = {
	-- dependent vowels (front type)
	["𑁂"] = "เ%1", ["𑁃"] = "ไ%1", ["𑁄"] = "โ%1", ["𑁅"] = "เ%1า",
}

function export.tr(text, lang, sc, debug_mode)

	if type(text) == "table" then -- called directly from a template
		text = text.args[1]
	end

	text = gsub(text, ".", tt)

	for k, v in pairs(adjust1) do
		text = gsub(text, letter_with_mark..k, v)
	end

	text = gsub(text, "([เแไโ])อฺ", "อฺ%1")

	-- ย้ายสัญลักษณ์ขึ้นบน เมื่อมีสระล่าง (ยกเว้นตัวที่ไม่มี)
	text = gsub(text, u(0x0331).."([ุ-ฺ])", u(0x0304).."%1") -- macron below > macron above

	return text

end

return export