Jump to content
Toggle search
Toggle menu
  • 4K articles
  • 2.9K files
  • 559 users
  • 36.8K edits
The Petit Planet Wiki
Toggle preferences menu
Toggle personal menu
Not logged in
If you make an edit, a temporary account will be created for you.

Module:Th-rm: Difference between revisions

From The Petit Planet Wiki
More actions
Browse history interactively

ReisuDesign (talk | contribs)

3 user groups
4,145 edits
undo
ReisuDesign (talk | contribs)
confirmed, Interface administrators, moderator
4,145 edits
Created page with "local p = {} local U = mw.ustring local gsub = U.gsub local find = U.find local sub = U.sub local match = U.match local len = U.len local INITIAL = { ['ก'] = 'k', ['ข'] = 'kh', ['ฃ'] = 'kh', ['ค'] = 'kh', ['ฅ'] = 'kh', ['ฆ'] = 'kh', ['ง'] = 'ng', ['จ'] = 'ch', ['ฉ'] = 'ch', ['ช'] = 'ch', ['ซ'] = 's', ['ฌ'] = 'ch', ['ญ'] = 'y', ['ฎ'] = 'd', ['ฏ'] = 't', ['ฐ'] = 'th', ['ฑ'] = 'th', ['ฒ'] = 'th', ['ณ'] = 'n', ['ด'] = 'd'..."
 
ReisuDesign (talk | contribs)
confirmed, Interface administrators, moderator
4,145 edits
No edit summary
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
local p = {}
local p = {}


local U = mw.ustring
local arguments = require('Module:Arguments')
local gsub = U.gsub
local tt = require('Module:Tt')
local find = U.find
local sub = U.sub
local match = U.match
local len = U.len


local INITIAL = {
local AUTO_TOOLTIP = 'This romanization was generated automatically. Please add a user-curated romanization as soon as possible.'
 
local THAI = '[ก-๛]'
 
-- Initial consonants.
local consonants = {
['ก'] = 'k',
['ก'] = 'k',
['ข'] = 'kh', ['ฃ'] = 'kh', ['ค'] = 'kh', ['ฅ'] = 'kh', ['ฆ'] = 'kh',
['ข'] = 'kh',
['ฃ'] = 'kh',
['ค'] = 'kh',
['ฅ'] = 'kh',
['ฆ'] = 'kh',
['ง'] = 'ng',
['ง'] = 'ng',
['จ'] = 'ch', ['ฉ'] = 'ch', ['ช'] = 'ch',
['จ'] = 'ch',
['ฉ'] = 'ch',
['ช'] = 'ch',
['ซ'] = 's',
['ซ'] = 's',
['ฌ'] = 'ch',
['ฌ'] = 'ch',
Line 18: Line 25:
['ฎ'] = 'd',
['ฎ'] = 'd',
['ฏ'] = 't',
['ฏ'] = 't',
['ฐ'] = 'th', ['ฑ'] = 'th', ['ฒ'] = 'th',
['ฐ'] = 'th',
['ฑ'] = 'th',
['ฒ'] = 'th',
['ณ'] = 'n',
['ณ'] = 'n',
['ด'] = 'd',
['ด'] = 'd',
['ต'] = 't',
['ต'] = 't',
['ถ'] = 'th', ['ท'] = 'th', ['ธ'] = 'th',
['ถ'] = 'th',
['ท'] = 'th',
['ธ'] = 'th',
['น'] = 'n',
['น'] = 'n',
['บ'] = 'b',
['บ'] = 'b',
Line 36: Line 47:
['ล'] = 'l',
['ล'] = 'l',
['ว'] = 'w',
['ว'] = 'w',
['ศ'] = 's', ['ษ'] = 's', ['ส'] = 's',
['ศ'] = 's',
['ษ'] = 's',
['ส'] = 's',
['ห'] = 'h',
['ห'] = 'h',
['ฬ'] = 'l',
['ฬ'] = 'l',
['ฮ'] = 'h'
['ฮ'] = 'h',
['อ'] = ''
}
}


local FINAL = {
-- Final consonants.
local finals = {
['ก'] = 'k',
['ก'] = 'k',
['ข'] = 'k', ['ฃ'] = 'k', ['ค'] = 'k', ['ฅ'] = 'k', ['ฆ'] = 'k',
['ข'] = 'k',
['ง'] = 'ng',
['ฃ'] = 'k',
['จ'] = 't', ['ฉ'] = 't', ['ช'] = 't', ['ฌ'] = '',
['ค'] = 'k',
['ฅ'] = 'k',
['ฆ'] = 'k',
 
['จ'] = 't',
['ฉ'] = 't',
['ช'] = 't',
['ซ'] = 't',
['ซ'] = 't',
['ญ'] = 'n',
['ฌ'] = 't',
['ฎ'] = 't', ['ฏ'] = 't',
['ฎ'] = 't',
['ฐ'] = 't', ['ฑ'] = 't', ['ฒ'] = 't',
['ฏ'] = 't',
['ณ'] = 'n',
['ฐ'] = 't',
['ฑ'] = 't',
['ฒ'] = 't',
['ด'] = 't',
['ด'] = 't',
['ต'] = 't',
['ต'] = 't',
['ถ'] = 't', ['ท'] = 't', ['ธ'] = 't',
['ถ'] = 't',
['น'] = 'n',
['ท'] = 't',
['ธ'] = 't',
['ศ'] = 't',
['ษ'] = 't',
['ส'] = 't',
 
['บ'] = 'p',
['บ'] = 'p',
['ป'] = 'p',
['ป'] = 'p',
['ผ'] = '', ['ฝ'] = '',
['ผ'] = 'p',
['พ'] = 'p', ['ฟ'] = 'p', ['ภ'] = 'p',
['ฝ'] = 'p',
['ม'] = 'm',
['พ'] = 'p',
['ย'] = '',
['ฟ'] = 'p',
['ภ'] = 'p',
 
['ง'] = 'ng',
 
['ญ'] = 'n',
['ณ'] = 'n',
['น'] = 'n',
['ร'] = 'n',
['ร'] = 'n',
['ล'] = 'n',
['ล'] = 'n',
['ว'] = '',
['ศ'] = 't', ['ษ'] = 't', ['ส'] = 't',
['ห'] = '',
['ฬ'] = 'n',
['ฬ'] = 'n',
['ฮ'] = ''
 
['ม'] = 'm'
}
}


-- Vowel patterns are tested longest-first.
-- Vowels.
local VOWELS = {
local vowels = {
{ 'เ' , 'ะ', 'e' },
['เียะ'] = 'ia',
{ 'เ' , '็', 'e' },
['เีย'] = 'ia',
{ 'เ' , '',  'e' },
['เือะ'] = 'uea',
['เือ'] = 'uea',
['เือย'] = 'ueai',
['เียว'] = 'iao',


{ 'แ' , 'ะ', 'ae' },
['ัวะ'] = 'ua',
{ 'แ' , '',  'ae' },
['ัว'] = 'ua',


{ 'โ' , 'ะ', 'o' },
['เิ็ย'] = 'oei',
{ 'โ' , '',  'o' },
['เิย'] = 'oei',
{ 'เ' , 'าะ', 'o' },
['เอย'] = 'oei',
{ ''  , 'อ', 'o' },


{ 'เ' , 'อะ', 'oe' },
['็อย'] = 'oi',
{ 'เ' , 'ิ', 'oe' },
['อย'] = 'oi',
{ 'เ' , 'อ', 'oe' },


{ 'เ' , 'ียะ', 'ia' },
['ุย'] = 'ui',
{ 'เ' , 'ีย', 'ia' },
['วย'] = 'uai',


{ 'เ' , 'ือะ', 'uea' },
['ัย'] = 'ai',
{ 'เ' , 'ือ', 'uea' },
['ไย'] = 'ai',
['าย'] = 'ai',
['ใ'] = 'ai',
['ไ'] = 'ai',


{ '' , 'ัวะ', 'ua' },
['เ็ว'] = 'eo',
{ '' , 'ัว', 'ua' },
['แ็ว'] = 'aeo',
{ '' , 'ว', 'ua' },
['เา'] = 'ao',
['าว'] = 'ao',
['แว'] = 'aeo',
['เว'] = 'eo',


{ 'ใ', '', 'ai' },
['ิว'] = 'io',
{ 'ไ', '', 'ai' },
['ีว'] = 'io',
{ '' , 'ัย', 'ai' },
{ 'ไ', 'ย', 'ai' },
{ '' , 'าย', 'ai' },


{ 'เ', 'า', 'ao' },
['ำ'] = 'am',
{ '' , 'าว', 'ao' },


{ '' , 'ุย', 'ui' },
['ือ'] = 'ue',
['ื'] = 'ue',
['ึ'] = 'ue',


{ 'โ', 'ย', 'oi' },
['ิ'] = 'i',
{ '' , 'อย', 'oi' },
['ี'] = 'i',


{ 'เ', 'ย', 'oei' },
['ุ'] = 'u',
{ 'เ', 'ือย', 'ueai' },
['ู'] = 'u',
{ '' , 'วย', 'uai' },


{ '' , 'ิว', 'io' },
['เ็'] = 'e',
['เ'] = 'e',


{ 'เ', '็ว', 'eo' },
['แ็'] = 'ae',
{ 'เ', 'ว', 'eo' },
['แ'] = 'ae',


{ 'แ', '็ว', 'aeo' },
['โะ'] = 'o',
{ 'แ', 'ว', 'aeo' },
['โ'] = 'o',


{ 'เ', 'ียว', 'iao' },
['เาะ'] = 'o',
['อ'] = 'o',


{ '' , 'ะ', 'a' },
['เอะ'] = 'oe',
{ '' , 'ั', 'a' },
['เอ'] = 'oe',
{ '' , 'า', 'a' },
['เะ'] = 'e',
{ '' , 'ำ', 'am' },


{ '' , 'ิ', 'i' },
['ะ'] = 'a',
{ '' , 'ี', 'i' },
['ั'] = 'a',
['า'] = 'a'
}


{ '' , 'ึ', 'ue' },
-- Longest vowel forms must be checked first.
{ '' , 'ื', 'ue' },
local vowel_order = {
{ '' , 'ือ', 'ue' },
'เือย',
'เียว',
'เือะ',
'เียะ',
'เีย',
'เือ',
'ัวะ',
'ัว',
'เิ็ย',
'เิย',
'็อย',
'เ็ว',
'แ็ว',
'เอย',
'เา',
'าว',
'เว',
'แว',
'าย',
'ไย',
'ัย',
'ุย',
'วย',
'อย',
'ำ',
'ิว',
'ีว',
'ิ',
'ี',
'ึ',
'ื',
'ุ',
'ู',
'เ็',
'แ็',
'เาะ',
'เอะ',
'เอ',
'เะ',
'ะ',
'ั',
'า',
'เ',
'แ',
'โ',
'อ',
'ใ',
'ไ'
}


{ '' , 'ุ', 'u' },
local consonant_set = {}
{ '' , 'ู', 'u' },


{ '' , 'รร', 'a' }
for c in pairs(consonants) do
consonant_set[c] = true
end
 
-- Thai consonant clusters.
local valid_clusters = {
['กร'] = true,
['กล'] = true,
['กว'] = true,
['ขร'] = true,
['ขล'] = true,
['ขว'] = true,
['คร'] = true,
['คล'] = true,
['คว'] = true,
['ฆร'] = true,
['ฆล'] = true,
['จร'] = true,
['ตร'] = true,
['ปร'] = true,
['ปล'] = true,
['พร'] = true,
['พล'] = true,
['ผล'] = true,
['ทร'] = true,
['ธร'] = true,
['ศร'] = true,
['สร'] = true,
['สล'] = true,
['สว'] = true,
['ซร'] = true,
['ฝร'] = true,
['ฟร'] = true
}
}
-- ห is silent in these clusters when followed by a vowel.
local h_silent_clusters = {
['หง'] = true,
['หน'] = true,
['หม'] = true,
['หย'] = true,
['หร'] = true,
['หล'] = true,
['หว'] = true,
['หญ'] = true
}
-- ร is silent in these clusters when followed by a vowel.
local r_silent_clusters = {
['จร'] = true,
['ซร'] = true,
['ศร'] = true,
['สร'] = true
}
local function chars(text)
local result = {}
for c in mw.text.gsplit(text, '') do
table.insert(result, c)
end
return result
end


local function isConsonant(c)
local function isConsonant(c)
return c and INITIAL[c] ~= nil
return c and consonant_set[c] == true
end
end


local function isThai(c)
local function isThai(c)
return c and match(c, '[ก-๛]') ~= nil
return c and mw.ustring.find(c, THAI) ~= nil
end
end


local function isToneMark(c)
local function isFrontVowel(c)
return c and match(c, '[่้๊๋]') ~= nil
return c == 'เ' or c == 'แ' or c == 'โ' or c == 'ใ' or c == 'ไ'
end
end


local function isSilentMark(c)
local function skipMarks(c, pos)
return c == '์'
while c[pos] and (
c[pos] == '็' or
c[pos] == '่' or
c[pos] == '้' or
c[pos] == '๊' or
c[pos] == '๋' or
c[pos] == 'ํ' or
c[pos] == '๎'
) do
pos = pos + 1
end
 
return pos
end
end


local function clean(str)
local function removeSilent(chars_)
str = mw.text.unstrip(str or '')
local result = {}
str = gsub(str, '<br%s*/?>', ' ')
local i = 1
str = gsub(str, '<[^>]+>', '')
 
return str
while chars_[i] do
if chars_[i] == '์' then
-- ์ makes the following consonant(s) silent.
i = i + 1
 
while chars_[i] and (
isConsonant(chars_[i]) or
chars_[i] == '็' or
chars_[i] == '่' or
chars_[i] == '้' or
chars_[i] == '๊' or
chars_[i] == '๋'
) do
i = i + 1
end
else
table.insert(result, chars_[i])
i = i + 1
end
end
 
return result
end
end


local function applyOrthographicExceptions(str)
local function getVowel(c, pos)
-- Silent letters marked with ์.
local remaining = table.concat(c, '', pos)
-- Remove the marked letter and its ์ marker.
str = gsub(str, '([ก-ฮ])์', '')


-- Tone marks have no RTGS representation.
for _, form in ipairs(vowel_order) do
str = gsub(str, '[่้๊๋]', '')
local length = mw.ustring.len(form)


-- ฤ
if mw.ustring.sub(remaining, 1, length) == form then
str = gsub(str, 'ฤกษ์', 'roe')
return form, vowels[form]
str = gsub(str, 'ฤิ', 'ri')
end
str = gsub(str, 'ฤ', 'rue')
end
str = gsub(str, 'ฤๅ', 'rue')


-- ฦ / ฦๅ
return nil, nil
str = gsub(str, 'ฦๅ', 'lue')
str = gsub(str, 'ฦ', 'lue')
 
return str
end
end


local function getVowel(str, pos)
local function getInitial(c, pos)
local remaining = sub(str, pos, -1)
local c1 = c[pos]


-- Explicit special vowel patterns.
if not isConsonant(c1) then
local patterns = {
return '', pos
{ 'เ', 'ื', 'อ', 'ย', 'ueai' },
end
{ 'เ', 'ี', 'ย', 'ว', 'iao' },
{ 'เ', 'ี', 'ย', 'ะ', 'ia' },
{ 'เ', 'ี', 'ย', '',  'ia' },
{ 'เ', 'ื', 'อ', 'ะ', 'uea' },
{ 'เ', 'ื', 'อ', '',  'uea' },


{ 'เ', '็', 'ว', '', 'eo' },
local c2 = c[pos + 1]
{ 'เ', 'ว', '', '', 'eo' },
{ 'แ', '็', 'ว', '', 'aeo' },
{ 'แ', 'ว', '', '', 'aeo' },


{ 'โ', 'ย', '', '', 'oi' },
-- ห is not pronounced in หง หน หม หย หร หล หว หญ.
{ '', 'อ', 'ย', '', 'oi' },
if c1 == 'ห' and c2 and h_silent_clusters[c1 .. c2] then
{ 'เ', 'ย', '', '', 'oei' },
return consonants[c2], pos + 2, true
end


{ '', 'ว', 'ย', '', 'uai' },
-- ทร is pronounced s in the relevant Sanskrit-derived words.
{ '', 'ุ', 'ย', '', 'ui' },
-- This is the broad automatic fallback; exceptions can be manually
{ '', 'ิ', 'ว', '', 'io' },
-- supplied with th_rm.
if c1 == 'ท' and c2 == 'ร' then
return 's', pos + 2, false
end


{ 'เ', 'า', '', '', 'ao' },
if c2 and valid_clusters[c1 .. c2] then
{ '', 'า', 'ว', '', 'ao' },
-- ร is silent in these clusters.
if r_silent_clusters[c1 .. c2] then
return consonants[c1], pos + 2, false
end


{ '', 'ั', 'ว', 'ะ', 'ua' },
return consonants[c1] .. consonants[c2], pos + 2, false
{ '', 'ั', 'ว', '', 'ua' },
end
{ '', 'ว', '', '', 'ua' },


{ 'ใ', '', '', '', 'ai' },
return consonants[c1], pos + 1, false
{ 'ไ', '', '', '', 'ai' },
end
{ '', 'ั', 'ย', '', 'ai' },
{ 'ไ', 'ย', '', '', 'ai' },
{ '', 'า', 'ย', '', 'ai' },


{ 'เ', 'า', 'ะ', '', 'ao' },
local function getCoda(c, pos)
{ 'โ', 'ะ', '', '', 'o' },
local current = c[pos]
{ 'เ', 'า', '', '', 'ao' },
{ 'เ', 'าะ', '', '', 'o' },


{ 'เ', 'อ', '', '', 'oe' },
if not current or not isConsonant(current) then
{ 'เ', 'ิ', '', '', 'oe' },
return '', pos
{ 'เ', 'อะ', '', '', 'oe' },
end


{ 'แ', 'ะ', '', '', 'ae' },
-- These do not act as ordinary codas here.
{ 'แ', '', '', '', 'ae' },
if current == 'อ' or current == 'ห' or current == 'ฮ' then
{ 'เ', 'ะ', '', '', 'e' },
return '', pos
{ 'เ', '็', '', '', 'e' },
end
{ 'เ', '', '', '', 'e' },
{ 'โ', '', '', '', 'o' },


{ '', 'ำ', '', '', 'am' },
-- ย and ว are generally part of the vowel/diphthong.
{ '', 'ะ', '', '', 'a' },
if current == 'ย' or current == 'ว' then
{ '', 'ั', '', '', 'a' },
return '', pos
{ '', 'า', '', '', 'a' },
end
{ '', 'ิ', '', '', 'i' },
{ '', 'ี', '', '', 'i' },
{ '', 'ึ', '', '', 'ue' },
{ '', 'ือ', '', '', 'ue' },
{ '', 'ื', '', '', 'ue' },
{ '', 'ุ', '', '', 'u' },
{ '', 'ู', '', '', 'u' },
{ '', 'อ', '', '', 'o' }
}


for _, v in ipairs(patterns) do
if finals[current] then
local a, b, c, d, roman = v
return finals[current], pos + 1
local pattern = a .. b .. c .. d
if pattern ~= '' and sub(remaining, 1, len(pattern)) == pattern then
return roman, len(pattern)
end
end
end


-- Implied a.
return '', pos + 1
return 'a', 0
end
end


local function nextVowelPosition(str, pos)
local function startsNextSyllable(c, pos)
local i = pos
if not c[pos] or not isConsonant(c[pos]) then
while i <= len(str) do
return false
local c = sub(str, i, i)
end
 
if isFrontVowel(c[pos + 1]) then
return true
end


if c == 'เ' or c == 'แ' or c == 'โ' or c == 'ใ' or c == 'ไ' then
local _, vowel = getVowel(c, pos + 1)
return i
 
end
return vowel ~= nil
end


if match(c, '[ะาิีึืุูัำอ]') then
local function parseSyllable(c, start)
return i
local pos = start
end


i = i + 1
if not c[pos] then
return nil, pos
end
end


return nil
local initial = ''
end
local cluster = false


local function romanizeSyllable(str, startPos)
-- Preposed vowels.
local pos = startPos
local front = ''
local n = len(str)


-- Preposed vowel.
if isFrontVowel(c[pos]) then
local preVowel = ''
front = c[pos]
local preLen = 0
local c = sub(str, pos, pos)
 
if c == 'เ' or c == 'แ' or c == 'โ' or c == 'ใ' or c == 'ไ' then
preVowel = c
preLen = 1
pos = pos + 1
pos = pos + 1
end
end


-- Initial consonant.
local initialPos = pos
local initial = sub(str, pos, pos)


if not isConsonant(initial) then
initial, pos, _, cluster =
-- Non-Thai character.
getInitial(c, pos)
return initial, 1
 
if initial == '' and not isConsonant(c[initialPos]) then
return nil, start + 1
end
end


pos = pos + 1
local vowel
local initialRom = INITIAL[initial] or ''
local vowelForm
 
if front ~= '' then
-- Handle the vowel forms which contain a preposed vowel.
if front == 'เ' and c[pos] == 'ี' and c[pos + 1] == 'ย' then
vowel = 'ia'
vowelForm = 'เีย'
pos = pos + 2
 
elseif front == 'เ' and c[pos] == 'ื' and c[pos + 1] == 'อ' then
vowel = 'uea'
vowelForm = 'เือ'
pos = pos + 2
 
elseif front == 'เ' and c[pos] == 'อ' then
vowel = 'oe'
vowelForm = 'เอ'
pos = pos + 1


-- ห- exception.
elseif front == 'เ' and c[pos] == 'า' then
if initial == 'ห' then
vowel = 'ao'
local second = sub(str, pos, pos)
vowelForm = 'เา'
if second and match(second, '[งนมยรลวญณฬ]') then
initialRom = INITIAL[second]
pos = pos + 1
pos = pos + 1
initial = second
 
elseif front == 'เ' and c[pos] == '็' then
vowel = 'e'
vowelForm = 'เ็'
pos = pos + 1
 
elseif front == 'แ' and c[pos] == '็' then
vowel = 'ae'
vowelForm = 'แ็'
pos = pos + 1
 
elseif front == 'โ' and c[pos] == 'ะ' then
vowel = 'o'
vowelForm = 'โะ'
pos = pos + 1
 
elseif front == 'ใ' or front == 'ไ' then
vowel = 'ai'
vowelForm = front
 
else
vowel = vowels[front]
vowelForm = front
end
else
vowelForm, vowel = getVowel(c, pos)
 
if vowelForm then
pos = pos + mw.ustring.len(vowelForm)
end
end
end
end


-- Consonant clusters where ร is not represented.
-- Implied vowels.
if (initial == 'จ' or initial == 'ซ' or initial == 'ศ' or initial == 'ส')
if not vowel then
and sub(str, pos, pos) == 'ร' then
if c[pos] and isConsonant(c[pos]) then
initialRom = INITIAL[initial]
if startsNextSyllable(c, pos) then
pos = pos + 1
vowel = 'a'
else
vowel = 'o'
end
else
vowel = 'a'
end
end
end


-- Find the vowel from the current position, including preposed vowels.
pos = skipMarks(c, pos)
local vowelStart = startPos + preLen + 1
local vowelRoman
local vowelLen


if preLen > 0 then
-- Final consonant.
vowelRoman, vowelLen = getVowel(str, startPos + 1)
local coda = ''
else
local codaPos = pos
vowelRoman, vowelLen = getVowel(str, pos)
end


-- If the vowel was explicitly consumed by getVowel, advance.
if c[pos] and isConsonant(c[pos]) then
if vowelLen > 0 then
-- If another consonant begins the next syllable, don't consume
if preLen > 0 then
-- it as a coda.
pos = startPos + preLen + 1 + vowelLen
if startsNextSyllable(c, pos) then
coda = ''
else
else
pos = pos + vowelLen
coda, codaPos = getCoda(c, pos)
end
end
end
end


-- Collect possible final consonant.
pos = codaPos
local final = sub(str, pos, pos)
pos = skipMarks(c, pos)
local finalRom = ''


if isConsonant(final) then
return {
local after = sub(str, pos + 1, pos + 1)
initial = initial,
vowel = vowel,
coda = coda,
next = pos
}, pos
end


-- A consonant followed by a vowel-starting syllable is normally
local function shouldHyphen(previous, current)
-- treated as the next syllable's initial rather than this syllable's final.
if not previous or not current then
local nextIsVowel = after and (
return false
match(after, '[ะาิีึืุูัำอ]') or
end
match(after, '[เแโใไ]')
)


if not nextIsVowel then
-- Before a succeeding syllable beginning with a vowel.
finalRom = FINAL[final] or ''
if current.initial == '' and current.vowel ~= '' then
pos = pos + 1
return true
 
-- -ร following another final consonant is silent in RTGS.
if final == 'ร' then
finalRom = 'n'
end
end
end
end


-- ทร- exception.
-- Before ng when the preceding syllable ends in a vowel.
if initial == 'ท' and sub(str, pos, pos) == 'ร' then
if previous.coda == '' and current.initial == 'ng' then
initialRom = 's'
return true
pos = pos + 1
end
end


return initialRom .. vowelRoman .. finalRom, math.max(1, pos - startPos)
return false
end
end


local function romanizeWord(word)
local function romanizeWord(word)
if word == '' then
local c = chars(word)
return ''
c = removeSilent(c)
end
 
local syllables = {}
local pos = 1
local safety = 0


-- Manual markers:
while pos <= #c and safety < 1000 do
-- < > = force space
safety = safety + 1
-- <-> = force hyphen
word = gsub(word, '< >', ' ')
word = gsub(word, '<->', '-')


local output = {}
if isConsonant(c[pos]) or isFrontVowel(c[pos]) then
local pos = 1
local syllable, nextPos = parseSyllable(c, pos)
local n = len(word)


while pos <= n do
if syllable then
local c = sub(word, pos, pos)
table.insert(syllables, syllable)


if c == '<' then
if nextPos <= pos then
local close = find(word, '>', pos + 1, true)
pos = pos + 1
if close then
else
pos = close + 1
pos = nextPos
end
else
else
table.insert(output, c)
-- Fail safely rather than looping forever.
pos = pos + 1
pos = pos + 1
end
end
elseif not isThai(c) then
else
table.insert(output, c)
-- Thai punctuation/marks that aren't part of a syllable.
pos = pos + 1
pos = pos + 1
else
local roman, consumed = romanizeSyllable(word, pos)
table.insert(output, roman)
pos = pos + consumed
end
end
end
local output = {}
for i, syllable in ipairs(syllables) do
if shouldHyphen(syllables[i - 1], syllable) then
table.insert(output, '-')
end
table.insert(
output,
(syllable.initial or '') ..
(syllable.vowel or '') ..
(syllable.coda or '')
)
end
end


Line 424: Line 626:
end
end


local function capitalize(str, sentence)
-- Parse the text while treating <> and <-> as hard boundaries.
local chars = mw.text.split(str, '')
--
local shouldCap = true
-- <>  = literal space
-- <-> = literal hyphen
--
-- Because each Thai section is romanized independently, automatic
-- hyphenation can never cross one of these markers.
local function romanizeThaiText(text)
local result = {}
local pos = 1
local length = mw.ustring.len(text)


for i, c in ipairs(chars) do
while pos <= length do
if match(c, '%a') then
-- Manual hyphen.
if shouldCap then
if mw.ustring.sub(text, pos, pos + 2) == '<->' then
chars[i] = U.upper(c)
table.insert(result, '-')
shouldCap = false
pos = pos + 3
elseif sentence then
 
shouldCap = false
-- Manual space.
elseif mw.ustring.sub(text, pos, pos + 1) == '<>' then
table.insert(result, ' ')
pos = pos + 2
 
-- Thai run.
elseif isThai(mw.ustring.sub(text, pos, pos)) then
local start = pos
 
while pos <= length do
local character = mw.ustring.sub(text, pos, pos)
 
if not isThai(character) then
break
end
 
-- Stop before a manual control marker.
if mw.ustring.sub(text, pos, pos + 2) == '<->' then
break
end
 
if mw.ustring.sub(text, pos, pos + 1) == '<>' then
break
end
 
pos = pos + 1
end
end
elseif c == ' ' then
 
if sentence then
local word = mw.ustring.sub(text, start, pos - 1)
shouldCap = true
 
else
table.insert(result, romanizeWord(word))
shouldCap = true
 
end
-- Everything else is preserved.
elseif match(c, '[%.!?]') then
else
shouldCap = true
table.insert(
result,
mw.ustring.sub(text, pos, pos)
)
 
pos = pos + 1
end
end
 
return table.concat(result)
end
 
local function convertThaiDigits(text)
local digits = {
['๐'] = '0',
['๑'] = '1',
['๒'] = '2',
['๓'] = '3',
['๔'] = '4',
['๕'] = '5',
['๖'] = '6',
['๗'] = '7',
['๘'] = '8',
['๙'] = '9'
}
 
for thai, latin in pairs(digits) do
text = mw.ustring.gsub(text, thai, latin)
end
 
return text
end
 
local function hasThai(text)
return mw.ustring.find(text, THAI) ~= nil
end
 
local function validateRomanization(text)
-- Never allow Thai characters to leak into the final result.
if hasThai(text) then
return false
end
 
return true
end
 
local function capitalizeWords(text)
return mw.ustring.gsub(
text,
'(%a)([%a]*)',
function(first, rest)
return mw.ustring.upper(first) .. rest
end
end
)
end
local function capitalizeSentence(text)
local prefix, first = mw.ustring.match(
text,
'^([%s%p]*)(%a)'
)
if not first then
return text
end
end


return table.concat(chars)
return prefix ..
mw.ustring.upper(first) ..
mw.ustring.sub(text, mw.ustring.len(prefix) + 2)
end
 
function p.romanize(text)
if not text or mw.text.trim(text) == '' then
return ''
end
 
-- Do not automatically romanize Latin-only text.
if not hasThai(text) then
return ''
end
 
text = convertThaiDigits(text)
 
local result = romanizeThaiText(text)
 
result = mw.ustring.gsub(result, '%s+', ' ')
result = mw.text.trim(result)
 
if not validateRomanization(result) then
return ''
end
 
return result
end
end


function p.main(frame)
function p.main(frame)
local args = require('Module:Arguments').getArgs(frame, {
local args = arguments.getArgs(frame, {
parentFirst = true,
parentFirst = true,
wrappers = {'Template:Th-rm'}
wrappers = {'Template:Th-rm'}
})
})


if not args[1] or args[1] == '' then
local text = args[1]
 
if not text or mw.text.trim(text) == '' then
return ''
return ''
end
end


local str = clean(args[1])
if not hasThai(text) then
 
-- Don't romanize text which contains no Thai script.
if not match(str, '[ก-๛]') then
return ''
return ''
end
end


str = applyOrthographicExceptions(str)
local result = p.romanize(text)


local result = romanizeWord(str)
if result == '' then
return ''
end


if args.capi then
if args.capi then
result = capitalize(result, false)
result = capitalizeWords(result)
elseif args.sent then
result = capitalize(result, true)
end
end


return result
if args.sent then
result = capitalizeSentence(result)
end
 
-- Automatically generated romanizations get a Tt warning.
return tt._main({
[1] = result,
[2] = AUTO_TOOLTIP,
['class'] = args.class
})
end
end


return p
return p
Retrieved from "https://petitplanet.wiki/Module:Th-rm"