Module:sa-Orya-translit

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

This module will transliterate Sanskrit language text per WT:SA TR. 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:sa-Orya-translit/testcases.

Functions

[edit]
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 consonants = {
	['କ']='k', ['ଖ']='kh', ['ଗ']='g', ['ଘ']='gh', ['ଙ']='ṅ',
	['ଚ']='c', ['ଛ']='ch', ['ଜ']='j', ['ଝ']='jh', ['ଞ']='ñ', 
	['ଟ']='ṭ', ['ଠ']='ṭh', ['ଡ']='ḍ', ['ଢ']='ḍh', ['ଣ']='ṇ', 
	['ତ']='t', ['ଥ']='th', ['ଦ']='d', ['ଧ']='dh', ['ନ']='n', 
	['ପ']='p', ['ଫ']='ph', ['ବ']='b', ['ଭ']='bh', ['ମ']='m',
	['ଯ']='y', ['ୟ']='y', ['ର']='r', ['ଲ']='l', ['ଳ']='ḷ', ['ଵ']='v',
	['ଶ']='ś', ['ଷ']='ṣ', ['ସ']='s', ['ହ']='h',
}

local diacritics = {
	['ା']='ā', ['ି']='i', ['ୀ']='ī', ['ୁ']='u', ['ୂ']='ū', ['ୃ']='ṛ', ['ୄ']='ṝ', 
	['ୢ']='ḷ', ['ୣ']='ḹ', ['େ']='e', ['ୈ']='ai', ['ୋ']='o', ['ୌ']='au',  ['୍']='',
}

local tt = {
	-- vowels
	['ଅ']='a', ['ଆ']='ā', ['ଇ']='i', ['ଈ']='ī', ['ଉ']='u', ['ଊ']='ū', ['ଋ']='ṛ', ['ୠ']='ṝ',
	['ଌ']='ḷ', ['ୡ']='ḹ', ['ଏ']='e', ['ଐ']='ai', ['ଓ']='o', ['ଔ']='au', 
	-- chandrabindu    
	['ଁ']='m̐', --until a better method is found
	-- anusvara    
	['ଂ']='ṃ', --until a better method is found
	-- visarga    
	['ଃ']='ḥ',
	-- avagraha
	['ଽ']='’',
	--numerals
	['૦']='୦', ['૧']='୧', ['૨']='୨', ['૩']='୩', ['૪']='୪', ['૫']='୫', ['૬']='୬', ['૭']='୭', ['૮']='୮', ['૯']='୯',
	--punctuation        
    ['॥']='.', --double danda
	['।']='.', --danda
    --Vedic extensions
    -- ['ᳵ']='x', ['ᳶ']='f',
    --Om
    ['ଓମ୍']='oṃ',
    --reconstructed
    ['*'] = '',
}

function export.tr(text, lang, sc)
	text = mw.ustring.gsub(
		text,
		'([କଖଗଘଙଚଛଜଝଞଟଠଡଢଣତଥଦଧନପଫବଭମଯୟରଲଳଵଶଷସହ])'..
		'([ାିୀୁୂୃୄୢୣେୈୋୌ୍]?)',
		function(c, d)
			if d == "" then        
				return consonants[c] .. 'a'
			else
				return consonants[c] .. diacritics[d]
			end
		end)

	text = mw.ustring.gsub(text, '.', tt)
	
	return text
end
 
return export