Module:oty-translit
Jump to navigation
Jump to search
- The following documentation is generated by Module:documentation/functions/translit. [edit]
- Useful links: subpage list โข links โข transclusions โข testcases โข sandbox
This module will transliterate Old Tamil 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:oty-translit/testcases.
Functions
[edit]tr(text, lang, sc)
- Transliterates a given piece of
text
written in the script specified by the codesc
, and language specified by the codelang
. - When the transliteration fails, returns
nil
.
local export = {}
local u = mw.ustring.char
local consonants = {
--consonants
['๐']='k', ['๐']='kh', ['๐']='g', ['๐']='gh', ['๐']='แน
',
['๐']='c', ['๐']='ch', ['๐']='j', ['๐']='jh', ['๐']='รฑ',
['๐']='แนญ', ['๐']='แนญh', ['๐']='แธ', ['๐ ']='แธh', ['๐ก']='แน',
['๐ข']='t', ['๐ฃ']='th', ['๐ค']='d', ['๐ฅ']='dh', ['๐ฆ']='n',
['๐ง']='p', ['๐จ']='ph', ['๐ฉ']='b', ['๐ช']='bh', ['๐ซ']='m',
['๐ฌ']='y', ['๐ญ']='r', ['๐ฎ']='l', ['๐ฏ']='v', ['๐ด']='แธท',
['๐ฐ']='ล', ['๐ฑ']='แนฃ', ['๐ฒ']='s', ['๐ณ']='h',
-- Old Tamil
[u(0x11075)] = 'แธท',
['๐ต']='แธป', ['๐ถ']='แน', ['๐ท'] ='แน',
}
local diacritics = {
--matras
['๐ธ']='ฤ', ['๐บ']='i', ['๐ป']='ฤซ', ['๐ผ']='u', ['๐ฝ']='ลซ', ['๐พ']='แน', ['๐ฟ']='แน',
['๐']='lฬฅ', ['๐']='lฬฅฬ', ['๐']='e', ['๐']='ai', ['๐']='o', ['๐
']='au', ['๐']='',
--bhattiprolu aa
['๐น']='ฤ',
--Old Tamil
['๐ณ']='ฤ', ['๐ด']='ล', [u(0x11070)]='',
-- Old Tamil up to and including Unicode 13.0
['๐๐']='ฤ', ['๐๐']='ล', -- Two character vowels!
}
local diatrema = {
['๐']='รฏ', ['๐']='รผ',
}
local tt = {
--vowels
['๐
']='a', ['๐']='ฤ', ['๐']='i', ['๐']='ฤซ', ['๐']='u', ['๐']='ลซ', ['๐']='แน', ['๐']='แน',
['๐']='lฬฅ', ['๐']='lฬฅฬ', ['๐']='e', ['๐']='ai', ['๐']='o', ['๐']='au',
['๐ฑ']='ฤ', ['๐ฒ']='ล', --Old Tamil
-- chandrabindu
['๐']='mฬ', --until a better method is found
-- anusvara
['๐']='แน', --until a better method is found
-- visarga
['๐']='แธฅ',
--numerals
['๐ฆ']='0', ['๐ง']='1', ['๐จ']='2', ['๐ฉ']='3', ['๐ช']='4', ['๐ซ']='5', ['๐ฌ']='6', ['๐ญ']='7', ['๐ฎ']='8', ['๐ฏ']='9',
--punctuation
['๐']='.', --danda
['๐']='.' --double danda
}
function export.tr(text, lang, sc)
if sc ~= "Brah" then
return nil
end
if lang == "pra" then -- Route contextually shortened Prakrit vowels through Old Tamil short vowels
text = mw.ustring.gsub(text, '(๐)([๐
-๐ณ]?)(๐)([๐
-๐ณ]?)', '๐ณ%2%3%4')
text = mw.ustring.gsub(text, '(๐)([๐
-๐ณ]?)(๐)([๐
-๐ณ]?)', '๐ฑ%2%3%4')
text = mw.ustring.gsub(text, '(๐)([๐
-๐ณ]?)(๐)([๐
-๐ณ]?)', '๐ด%2%3%4')
text = mw.ustring.gsub(text, '(๐)([๐
-๐ณ]?)(๐)([๐
-๐ณ]?)', '๐ฒ%2%3%4')
end
text = mw.ustring.gsub(
text,
'([๐-๐ท'..u(0x11075)..'])'..
'([๐ธ๐บ๐บ๐ป๐ผ๐ฝ๐พ๐ฟ๐๐๐๐๐๐
๐๐น๐ณ๐ด'..u(0x11070)..']?๐?)'..
'([๐๐]?)',
function(c, d, e)
if d == "" and e ~= "" then
return consonants[c] .. 'a' .. diatrema[e]
elseif e ~= "" then
return consonants[c] .. diacritics[d] .. tt[e]
elseif d == "" then
return consonants[c] .. 'a'
else
return consonants[c] .. diacritics[d]
end
end)
-- Adjacent vowel letters needing dieresis
text = mw.ustring.gsub(text, '([๐
])([๐๐])', function(a, b) return tt[a]..diatrema[b] end)
text = mw.ustring.gsub(text, '.', tt)
if (lang == 'sa' or lang == 'pi') and mw.ustring.match(text, 'lฬฅ') then
text = mw.ustring.gsub(text, 'lฬฅ', 'แธท')
text = mw.ustring.toNFC(text)
end
-- Old Tamil uses macron v. plain for 'e' and 'o'.
if (lang == 'oty') then
text = mw.ustring.gsub(text, '.', {e='ฤ', o='ล', ['ฤ']='e', ['ล']='o'})
end
return text
end
return export