Module:xsr-Deva-translit

From Wiktionary, the free dictionary
Jump to navigation Jump to search

This module will transliterate Sherpa language text. 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:xsr-Deva-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 match = mw.ustring.match

local conv = {
-- consonants

['क'] = 'k', ['ख'] = 'kh', ['ग'] = 'g', ['ङ'] = 'ng', 
['च'] = 'ts', ['छ'] = 'tsh', ['ज'] = 'dz', ['ञ'] = 'ny', 
['ट'] = 'ṭ', ['ठ'] = 'ṭh', ['ड'] = 'ḍ',  
['त'] = 't', ['थ'] = 'th', ['द'] = 'd', ['न'] = 'n', 
['प'] = 'p', ['फ'] = 'ph', ['ब'] = 'b', ['म'] = 'm', 
['य'] = 'y', ['र'] = 'r', ['ल'] = 'l', ['व'] = 'w', ['ळ'] = 'ḷ', 
['श'] = 'sh', ['स'] = 's', ['ह'] = 'h', 

['श़'] = 'zh', ['स़'] = 'z', ['ॽ'] = "'", 
['च़'] = 'c', ['छ़'] = 'ch',


 --
['घ'] = 'gh', ['झ'] = 'dzh',
['ढ'] = 'ḍh', ['ण'] = 'ṇ', ['ध'] = 'dh',
['भ'] = 'bh',  ['ष'] = 'ṣ', 
['क़'] = 'q', ['ख़'] = 'x', ['ग़'] = 'ġ', 
['ऴ'] = 'ḻ', ['ज़'] = 'z', ['झ़'] = 'ž', 
['ड़'] = 'ṛ', ['ढ़'] = 'ṛh', ['फ़'] = 'f', 
['थ़'] = 'θ',['द़'] = 'ð', ['ऩ'] = 'ṉ', 
['ऱ'] = 'ṟ', ['व़'] = 'V', 
['ॹ'] = 'ž', 
																																																																																																																																																																																																																																																																																																																																																																-- vowel diacritics
['ि'] = 'i', ['ु'] = 'u', ['े'] = 'e', ['ो'] = 'o', ['ा'] = 'ā', ['ी'] = 'ī', ['ू'] = 'ū', ['ृ'] = 'r̥', ['ॄ'] = 'r̥̄', ['ॢ'] = 'l̥', ['ॣ'] = 'l̥̄', ['ै'] = 'ai', ['ौ'] = 'au', ['ॉ'] = 'ŏ', ['ॅ'] = 'ĕ', 
																																																																														-- vowel signs
['अ'] = 'a', ['इ'] = 'i', ['उ'] = 'u', ['ए'] = 'e', ['ओ'] = 'o', ['आ'] = 'ā', ['ई'] = 'ī', ['ऊ'] = 'ū', ['ऋ'] = 'r̥', ['ॠ'] = 'r̥̄', ['ऌ'] = 'l̥', ['ॡ'] = 'l̥̄', ['ऐ'] = 'ai', ['औ'] = 'au', ['ऑ'] = 'ŏ', ['ॲ'] = 'ĕ', ['ऍ'] = 'ĕ', 
																																																																														-- chandrabindu
['ँ'] = '̃',
-- anusvara
['ं'] = '̃',
-- visarga
['ः'] = 'ḥ',
-- virama
['्'] = '',

-- numerals
['०'] = '0', ['१'] = '1', ['२'] = '2', ['३'] = '3', ['४'] = '4', ['५'] = '5', ['६'] = '6', ['७'] = '7', ['८'] = '8', ['९'] = '9', 
-- punctuation
['।'] = '.', -- danda
['॥'] = '.', -- double danda
['+'] = '', -- compound separator

-- abbreviation sign
['॰'] = '.',
		}
				local perm_cl = {
																							['ज्ञ'] = true,
																							}																

local all_cons, special_cons = 'कखगघङचछजझञटठडढणतथदधनपफबभमयरलवसशषह', 'कखगघचछजझटठडढणतथदधनपफबभमयरलवशषसह'
local vowel, vowel_sign = 'aिुृेोाीूैौॉॅॆॊॄॢॣ',  'अइउएओआईऊऋॠॡऌऐऔऑऍ'

local function rev_string(text)
	local result, length = {}, mw.ustring.len(text)
			for i = length, 1, -1 do
							table.insert(result, mw.ustring.sub(text, i, i))
												end
																		return table.concat(result)
																								end
																								function export.tr(text, lang, sc)
																									text = gsub(text, '([' .. all_cons .. ']़?)([' .. vowel .. '्]?)', function(c, d)
																												return c .. (d == "" and 'a' or d) end)
for word in mw.ustring.gmatch(text, "[ऀ-ॿa]+") do
																																						local orig_word = word
																																														word = rev_string(word)
																																																								word = gsub(word, '^a(़?)([' .. all_cons .. '])(.)(.?)', function(opt, first, second, third)
																																																																					return (((match(first, '[' .. special_cons .. ']') and match(second, 'ं')
or match(first, '[' .. special_cons .. ']') and match(second, '्') and not perm_cl[first..second..third])
																																																																																						or match(first .. second, 'य[aिुृेोाीूैौॉॅॆॊआईउऊइएऐओऔ]') or match(first .. second, 'ह[अaिुृेोाीूैौॉॅॆॊआईउऊइएऐओऔ]'))																																																																													
																																																																																																											and 'a' or "") .. opt .. first .. second .. third end)

word = gsub(word, '(.?)ं(.)', function(succ, prev)
																																																																																																																																																																																																																																																							return succ .. (succ..prev == "" and "्म" or 
																																																																																																																																																																																																																																																																																												(succ == "" and match(prev, '[' .. vowel .. ']') and "̃" or nasal_assim[succ] or "ṃ")) .. prev end)

local escaped_orig_word = gsub(orig_word, "%+", "")																																																																																																																																																																																																																																																																																																																																	text = gsub(text, orig_word, rev_string(word))	
text = gsub(text, "च्य", "c")
text = gsub(text, "छ्य", "ch")
text = gsub(text, "ज्य", "j")	
text = gsub(text, "झ्य", "jh")
text = gsub(text, "ङ्य", "ny")
text = gsub(text, "श्य", "zh")																																																																																																																																																																																																																																																																																																																																						end
																																																																																																																																																																																																																																																																																																																																																																																																																				text = gsub(text, '.़?', conv)
																																																																																																																																																																																																																																																																																																																																																																																																																																																														text = gsub(text, 'a([iu])̃', 'a͠%1')
text = gsub(text, "[<>]", "")
					    																																																																																																																																																																																																																																																																																																																																																																																																																																																																																																							    																																										        																																										        	return mw.ustring.toNFC(text)
																																																																																																																																																																																																																																																																																																																																																																																																																																																																																																								    																																										        																																										        																																											        	end
																																																																																																																																																																																																																																																																																																																																																																																																																																																																																																								    																																										        																																										        																																											        	return export