Module:jje-pron
- The following documentation is located at Module:jje-pron/documentation. [edit] Categories were auto-generated by Module:module categorization. [edit]
- Useful links: subpage list • links • transclusions • testcases • sandbox
Korean pronunciation and romanisation module. Generates Revised Romanisation (RR), phonetic hangul (ph), Yale romanisation (yr), WT-revised Revised Romanisation (rrr) and IPA phonemic (ipa-p) & IPA phonetic (ipa) for now.
Data is stored at Module:jje-pron/data.
See {{jje-IPA}}.
local export = {}
local m_data = mw.loadData("Module:jje-pron/data")
local m_jje_utilities = require("Module:jje")
local m_str_utils = require("Module:string utilities")
-- Created from Module:ko.pron
local codepoint = m_str_utils.codepoint
local concat = table.concat
local floor = math.floor
local gmatch = m_str_utils.gmatch
local gsub = m_str_utils.gsub
local insert = table.insert
local len = m_str_utils.len
local match = m_str_utils.match
local pattern_escape = m_str_utils.pattern_escape
local remove = table.remove
local sub = m_str_utils.sub
local toNFC = mw.ustring.toNFC
local u = m_str_utils.char
local upper = m_str_utils.upper
local PAGENAME = mw.loadData("Module:headword/data").pagename
local system_lookup = {
["ph"] = 1, ["rr"] = 2, ["rrr"] = 3,
["yr"] = 4, ["ipa-p"] = 5, ["ipa"] = 6,
}
local question_mark = "<sup><small>[[Wiktionary:About Korean/Romanization|?]]</small></sup>"
local system_list = {
{
abbreviation = "ph",
display = "Phonetic hangul: ",
separator = "/",
},
{
abbreviation = "rr",
display = "Revised Romanization" .. question_mark,
separator = "/",
},
{
abbreviation = "rrr",
display = "Revised Romanization (translit.)" .. question_mark,
separator = "/"
},
{
abbreviation = "yr",
display = "Yale Romanization" .. question_mark,
separator = "/"
},
{
abbreviation = "ipa-p",
display = "(<i>Morphophonemic</i>) [[Wiktionary:International Phonetic Alphabet|IPA]]<sup>([[w:Jeju language|key]])</sup>: ",
separator = " ~ "
},
{
abbreviation = "ipa",
display = "(<i>[[w:Jeju language#Geographic distribution|Jeju City]]</i>) [[Wiktionary:International Phonetic Alphabet|IPA]]<sup>([[w:Jeju language|key]])</sup>: ",
separator = " ~ "
}
}
--[[
vowel_variation:
rules for vowel transformation.
key:
the number of a syllable's vowel (vowel_id):
floor(((codepoint('가') - 0xAC00) % 588) / 28) = 0
floor(((codepoint('개') - 0xAC00) % 588) / 28) = 1
value:
an integer that is added to the decimal codepoint of the syllable
u(codepoint('개') + 112) = '게'
allowed_vowel_scheme:
a list of which systems vowel transformation is reflected in.
key:
vowel_id .. "-" .. system_index
system_index: see system_list above. IPA is #6
value:
1, representing true
]]
local final_syllable_conversion = { [""] = "Ø", ["X"] = "" }
local com_mc = { ["g"] = "k", ["d"] = "t", ["b"] = "p", ["j"] = "ch", ["sy"] = "s", ["s"] = "ss" }
local com_ph = { ["ᄀ"] = "ᄁ", ["ᄃ"] = "ᄄ", ["ᄇ"] = "ᄈ", ["ᄉ"] = "ᄊ", ["ᄌ"] = "ᄍ" }
local vowel_variation = {
[1] = 112, -- 개→게
[7] = -56, -- 계→게
}
local allowed_vowel_scheme = {
["1-1"] = 1,
["1-6"] = 1,
["7-1"] = 1,
["7-6"] = 1,
}
local ambiguous_intersyllabic_rr = { ["oe"] = 1, ["eo"] = 1, ["eu"] = 1, ["ae"] = 1, ["ui"] = 1 }
local ambiguous_intersyllabic_mr = { ["oe"] = 1, ["ae"] = 1 }
local function decompose_syllable(word)
local decomposed_syllables = {}
for syllable in mw.text.gsplit(word, "") do
insert(decomposed_syllables, m_jje_utilities.decompose_jamo(syllable))
end
return decomposed_syllables
end
local function tidy_phonetic(original, romanised)
local ori_index, rom_index, result = 1, 1, {}
for i = 1, len(romanised) do
local romanised_syllable = sub(romanised, rom_index, rom_index)
local original_syllable = sub(original, ori_index, ori_index)
if romanised_syllable ~= original_syllable then
-- Handle cases of non-precomposed choseong character changing
if match(romanised_syllable, "[ᄀ-ᄒ]") then
-- Insert choseong
insert(result, '<b>'..romanised_syllable..'</b>')
-- Insert jungseong
romanised_syllable = sub(romanised, rom_index + 1, rom_index + 1)
insert(result, '<b>'..romanised_syllable..'</b>')
-- Check if next character is jongseong; if so, insert, if not end iteration
-- and continue accoringly
romanised_syllable = sub(romanised, rom_index + 2, rom_index + 2)
if match(romanised_syllable, "ᆨ-ᇝ") then
insert(w, '<b>'..romanised_syllable..'</b>')
ori_index, rom_index = ori_index + 2, rom_index + 2
if match(original_syllable, "[^ ]") then rom_index = rom_index + 1 end
if match(romanised_syllable, "[^ ]") then ori_index = ori_index + 1 end
else
ori_index, rom_index = ori_index + 1, rom_index + 1
if match(original_syllable, "[^ ]") then rom_index = rom_index + 1 end
if match(romanised_syllable, "[^ ]") then ori_index = ori_index + 1 end
end
-- Handle cases of non-precomposed jongseong character changing
elseif match(romanised_syllable, "[ᆨ-ᇝ]") then
-- Remove previous non-bold choseong & jungseong
remove(result, #result - 1)
remove(result, #result)
-- Insert choseong character
romanised_syllable_cho = sub(romanised, rom_index - 2, rom_index - 2)
insert(result, '<b>'..romanised_syllable_cho..'</b>')
-- Insert jungseong character
romanised_syllable_jung = sub(romanised, rom_index - 1, rom_index - 1)
insert(result, '<b>'..romanised_syllable_jung..'</b>')
-- Insert jongseong character
insert(result, '<b>'..romanised_syllable..'</b>')
if match(original_syllable, "[^ ]") then rom_index = rom_index + 1 end
if match(romanised_syllable, "[^ ]") then ori_index = ori_index + 1 end
-- Handle precomposed characters per typical method
else
insert(result, '<b>'..romanised_syllable..'</b>')
if match(original_syllable, "[^ ]") then rom_index = rom_index + 1 end
if match(romanised_syllable, "[^ ]") then ori_index = ori_index + 1 end
end
else
insert(result, '<span>'..romanised_syllable..'</span>')
ori_index, rom_index = ori_index + 1, rom_index + 1
end
end
return concat(result)
end
local function tidy_ipa(ipa)
ipa = gsub(ipa, "ʌ̹ː", "ɘː")
ipa = gsub(ipa, "ɭɭ([ji])", "ʎʎ%1")
ipa = gsub(ipa, "s([ʰ͈])ɥi" ,"ʃ%1ɥi")
ipa = gsub(ipa, "ss͈([ji])" ,"ɕɕ͈%1")
ipa = gsub(ipa, "s([ʰ͈])([ji])" ,"ɕ%1%2")
ipa = gsub(ipa, "nj", "ɲj")
ipa = gsub(ipa, "([ʑɕ])([ʰ͈]?)j", "%1%2")
ipa = gsub(ipa, "kʰ[ijɯ]", {
["kʰi"] = "kçi",
["kʰj"] = "kçj",
["kʰɯ"] = "kxɯ" }
)
ipa = gsub(ipa, "[hɦ][ijɯouw]", {
["hi"] = "çi",
["hj"] = "çj",
["hɯ"] = "xɯ",
["ho"] = "ɸʷo",
["hu"] = "ɸʷu",
["hw"] = "ɸw",
["ɦi"] = "ʝi",
["ɦj"] = "ʝj",
["ɦɯ"] = "ɣɯ",
["ɦo"] = "βo",
["ɦu"] = "βu",
["ɦw"] = "βw" }
)
if match(ipa, "ɥi") then
local midpoint = floor(len(ipa) / 2)
ipa = sub(ipa, 1, midpoint) .. gsub(sub(ipa, midpoint+1, -1), "ɥi", "y")
end
return ipa
end
function export.romanise(text_param, system_index, args)
if type(text_param) == "table" then
args = text_param:getParent().args
system_index = args[2] or 2
text_param = args[1]
end
local p, optional_params = {}, { "nn", "l", "com", "cap", "ni" }
for _, pm in ipairs(optional_params) do
p[pm] = { }
if args[pm] then
for pp in mw.text.gsplit(args[pm], ",") do p[pm][tonumber(pp) or pp] = 1 end
end
end
local vowel_ui_i, vowel_ui_e, no_batchim, batchim_reduce, s_variation, iotation, yeo_reduce =
args.ui, args.uie, args.nobc, args.bcred, args.svar, args.iot, args.yeored
system_index = system_lookup[system_index] or system_index
text_param = gsub(text_param, '["](.)', "%1")
for primitive_word in gmatch(text_param, "[%-ᄀ-ᄒ".."ᅡ-ᆢ".."ᆨ-ᇝ" .. "ퟍ" .. "ᆦힲ" .. " ㄱ-ㆎ가-힣' -]+") do
local the_original = primitive_word
primitive_word = gsub(primitive_word, "'''", "ß")
local formatting_position, formatting_count = {}, 0
primitive_word = gsub(primitive_word, "()([ß-])", function(m1, m2)
formatting_position[m1 + formatting_count] = m2 == "ß" and "'''" or m2
return ""
end)
local has_vowel = {}
for ch in gmatch(primitive_word, ".") do
local jungseong = floor(((codepoint(ch) - 0xAC00) % 588) / 28)
if not match(ch, "[예옛례롄]") and match(ch, "[가-힣]") then has_vowel[jungseong] = true end
end
local word_set = { primitive_word }
local function add_respelling(variable, modification, modification2)
modification2 = modification2 or function(x) return x end
if variable and match(system_index, "[16]") then
variable = tonumber(variable)
local pre_length = #word_set
for i = 1, pre_length do
local item = mw.text.split(word_set[i], "")
item[variable] = modification(item[variable])
item[variable + 1] = modification2(item[variable + 1])
word_set[pre_length + i] = concat(item)
end
end
end
add_respelling(vowel_ui_i, function(x) return "이" end)
add_respelling(vowel_ui_e, function(x) return "에" end)
add_respelling(no_batchim,
function(x) return u(codepoint(x) - (codepoint(x) - 0xAC00) % 28) end,
function(y) return u(codepoint(y) + 588) end)
add_respelling(s_variation, function(x) return u(codepoint(x) - 12) end)
add_respelling(iotation, function(x) return u(codepoint(x) + 56) end)
add_respelling(yeo_reduce, function(x) return u(codepoint(x) - 56) end)
for vowel_id, vowel_variation_increment in pairs(vowel_variation) do
if has_vowel[vowel_id] and allowed_vowel_scheme[vowel_id .. "-" .. system_index] then
local pre_length = #word_set
for i = 1, pre_length do
local item = mw.text.split(word_set[i], "")
for num, it in ipairs(item) do
if floor(((codepoint(it) - 0xAC00) % 588) / 28) == vowel_id then
item[num] = u(codepoint(it) + vowel_variation_increment)
end
end
if vowel_id == 11 then
insert(word_set, i, concat(item))
else
insert(word_set, concat(item))
end
end
end
end
local word_set_romanisations = {}
for _, respelling in ipairs(word_set) do
local decomposed_syllables = decompose_syllable(respelling)
local romanisation = {}
local formatting_insert_count = 0
for index = 0, #decomposed_syllables, 1 do
local this_syllable_text = index ~= 0 and sub(respelling, index, index) or ""
if this_syllable_text == "-" then
-- skip it, it will be handled below
else
local syllable = decomposed_syllables[index] or { initial = "Ø", vowel = "Ø", final = "X" }
local next_index = index
local next_syllable_text
local saw_hyphen_after = false
while true do
next_index = next_index + 1
next_syllable_text = next_index > #decomposed_syllables and "" or sub(respelling, next_index, next_index)
if next_syllable_text ~= "-" then
break
end
saw_hyphen_after = true
end
local next_syllable = decomposed_syllables[next_index] or { initial = "Ø", vowel = "Ø", final = "Ø" }
syllable.final = final_syllable_conversion[syllable.final] or syllable.final
if system_index == 4 and syllable.vowel == "ᅮ" and match(syllable.initial, "[ᄆᄇᄈᄑ]") then
syllable.vowel = "ᅳ"
end
if match(system_index, "[126]") then
if match(syllable.initial, "[ᄌᄍᄎ]") then
if syllable.vowel == "ᅣ" then
syllable.vowel = "ᅡ"
elseif syllable.vowel == "ᅤ" then
syllable.vowel = "ᅢ"
elseif syllable.vowel == "ᅧ" then
syllable.vowel = "ᅥ"
elseif syllable.vowel == "ᅨ" then
syllable.vowel = "ᅦ"
elseif syllable.vowel == "ᅭ" then
syllable.vowel = "ᅩ"
elseif syllable.vowel == "ᅲ" then
syllable.vowel = "ᅮ"
elseif syllable.vowel == "ᆢ" then
syllable.vowel = "ᆞ"
end
end
end
if match(system_index, "[16]") then
if syllable.vowel == "ᅴ" and this_syllable_text ~= "의" then
syllable.vowel = "ᅵ"
end
end
if match(system_index, "[126]") then
if this_syllable_text == "넓" then
if match(next_syllable.initial, "[ᄌᄉ]") then
syllable.final = "ᆸ"
elseif next_syllable.initial == "ᄃ" then
if match(next_syllable.vowel, "[^ᅡᅵ]") then
syllable.final = "ᆸ"
end
end
end
end
local vowel = m_data.vowels[syllable.vowel][system_index]
if p.nn[next_index] then
next_syllable.initial = "ᄂ"
end
if p.com[index] and match(system_index, "[16]") then
next_syllable.initial = com_ph[next_syllable.initial] or next_syllable.initial
end
if p.ni[next_index] and system_index ~= 3 then
next_syllable.initial = (system_index == 4 and syllable.final == "ᆯ") and "ᄅ" or "ᄂ"
end
if match(system_index, "[^34]") then
if tonumber(batchim_reduce or -1) == index then
syllable.final = m_data.boundary[syllable.final .. "-Ø"][1]
end
if index ~= 0 and this_syllable_text == "밟" then
syllable.final = "ᆸ"
end
if match(syllable.final .. next_syllable.initial .. next_syllable.vowel, "[ᇀᆴ][ᄋᄒ]ᅵ") then
syllable.final = "ᆾ"
elseif match(syllable.final .. next_syllable.initial .. next_syllable.vowel, "ᆮ이") then
syllable.final = "ᆽ"
elseif match(syllable.final .. next_syllable.initial .. next_syllable.vowel, "ᆮ히") then
syllable.final = "ᆾ"
elseif match(syllable.final .. next_syllable.initial .. next_syllable.vowel, "[ퟍ][ㅇㅎ]ㅣ") then
syllable.final = "ퟹ"
elseif syllable.final .. next_syllable.initial == "ᆺᄋ" and not
match(next_syllable_text, "[아앗야어엇에예으은을음읍의이인일임입잇]") then
syllable.final = "ᆮ"
end
end
local bound = syllable.final .. "-" .. next_syllable.initial
if not m_data.boundary[bound] then
require("Module:debug").track("jje-pron/no boundary data")
mw.log("No boundary data for " .. bound .. ".")
return nil
end
local junction = m_data.boundary[bound][system_index]
if formatting_position[index + formatting_insert_count + 1] and system_index == 2 then
junction = gsub(junction, "^.*$", function(matched)
local a, b = match(matched, "^(ng%-?)(.?)$")
if not a or not b then a, b = match(matched, "^(.?%-?)(.*)$") end
return match(syllable.final .. next_syllable.initial, "^Ø?[ᄀ-ᄒ]$")
and formatting_position[index + formatting_insert_count + 1] .. (a or "") .. (b or "")
or (a or "") .. formatting_position[index + formatting_insert_count + 1] .. (b or "") end)
formatting_insert_count = formatting_insert_count + 1
end
if p.l[index] or (p.l["y"] and index == 1) then
-- FIXME, verify this code still works with final/initial cons changes
if system_index == 1 then
if #junction == 0 then
junction = junction .. "ː"
else
junction = gsub(junction, "^(.)(.?)$", function(a, b)
return match(a, "[ᆨ-ᇂ]") and a .. "ː" .. b or "ː" .. a .. b end)
end
elseif system_index == 5 then
vowel = gsub(vowel, "([aeiou])", "%1̄")
elseif system_index == 6 then
vowel = vowel .. "ː[[Category:Korean terms with long vowels in the first syllable]]"
end
end
if (p.l["y"] or p.l[1]) and index == 0 and system_index == 6 and #decomposed_syllables > 1 then
vowel = vowel .. "ˈ"
end
if p.com[index] then
-- FIXME, verify this code still works with final/initial cons changes
junction = gsub(junction, "(.)$", function(next_letter)
return
(system_index == 5 and "q" or "") ..
(system_index == 4
and (com_mc[next_letter..(p.cap["y"] or "")] or com_mc[next_letter] or next_letter)
or next_letter) end)
end
if p.ni[next_index] and system_index == 4 then
-- FIXME, verify this code still works with final/initial cons changes
junction = gsub(junction, "([nl])$", "<sup>%1</sup>")
end
local final_cons, initial_cons = match(junction, "^(.*);(.*)$")
if not final_cons then
if system_index == 2 then
error("Need a semicolon in the boundary value for " .. bound)
end
-- FIXME, throw an error for all systems once we've added semicolons everywhere
final_cons = junction
initial_cons = ""
end
if system_index == 2 then
insert(romanisation, vowel .. final_cons .. (saw_hyphen_after and "-" or "") .. initial_cons)
else
insert(romanisation, vowel .. junction)
end
end
end
local temp_romanisation = concat(romanisation)
if p.cap["y"] and match(system_index, "[^16]") then
temp_romanisation = upper(sub(temp_romanisation, 1, 1)) .. sub(temp_romanisation, 2, -1)
end
if system_index == 1 then
temp_romanisation = tidy_phonetic(primitive_word, toNFC(temp_romanisation))
elseif match(system_index, "[23]") then
for i = 1, 2 do
temp_romanisation = gsub(temp_romanisation, "(.)…(.)", function(a, b)
return a .. (ambiguous_intersyllabic_rr[a .. b] and "'" or "") .. b end)
temp_romanisation = gsub(temp_romanisation, "wo'e", "woe")
temp_romanisation = gsub(temp_romanisation, "yo'e", "yoe")
temp_romanisation = gsub(temp_romanisation, "we'o", "weo")
temp_romanisation = gsub(temp_romanisation, "we'u", "weu")
temp_romanisation = gsub(temp_romanisation, "ye'u", "yeu")
temp_romanisation = gsub(temp_romanisation, "yu'i", "yui")
end
elseif system_index == 5 then
temp_romanisation = "⫽" .. temp_romanisation .. "⫽"
elseif system_index == 6 then
temp_romanisation = "[" .. temp_romanisation .. "]"
end
insert(word_set_romanisations, temp_romanisation)
end
text_param = gsub(
text_param,
pattern_escape(the_original),
concat(word_set_romanisations, system_list[system_index].separator),
1
)
end
if system_index == 6 then
text_param = tidy_ipa(text_param)
end
return text_param
end
function export.make(frame, scheme)
local params = {
[1] = { default = PAGENAME, list = true },
["a"] = {},
["audio"] = { alias_of = "a" },
["nn"] = {},
["l"] = {},
["com"] = {},
["cap"] = {},
["ui"] = {},
["uie"] = {},
["nobc"] = {},
["ni"] = {},
["bcred"] = {},
["svar"] = {},
["iot"] = {},
["yeored"] = {},
}
local args = require("Module:parameters").process(frame:getParent().args, params)
local results = {}
for _, text_param in ipairs(args[1]) do
local current_word_dataset = {}
for system_index, system in pairs(system_list) do
local romanised = export.romanise(text_param, system_index, args)
insert(current_word_dataset, romanised)
end
insert(results, current_word_dataset)
end
local output_result = { [1] = {}, [2] = {}, [3] = {}, [4] = {}, [5] = {}, [6] = {} }
for _, result in ipairs(results) do
for result_index, value in ipairs(result) do
insert(output_result[result_index], value)
end
end
local html_ul = mw.html.create( "ul" )
:done()
local html_li_ipa_p = mw.html.create( "li" )
:wikitext( system_list[5].display )
:tag( "span" )
:addClass( "IPA" )
:wikitext( concat(output_result[5], system_list[5].separator) )
:done()
:done()
local html_li_ipa = mw.html.create( "li" )
:wikitext( system_list[6].display )
:tag( "span" )
:addClass( "IPA" )
:wikitext( concat(output_result[6], system_list[6].separator) )
:done()
:done()
local html_li_ph = mw.html.create( "li" )
:addClass( "jje-pron__ph" )
:wikitext( system_list[1].display )
:tag( "span" )
:addClass( "Kore" )
:attr( "lang", "jje" )
:wikitext( "[" .. concat(output_result[1], system_list[1].separator) .. "]" )
:done()
:done()
if args.a then
html_li_ipa
:tag( "ul" )
:tag( "li" )
:wikitext( require("Module:audio").format_audio {
lang = require("Module:languages").getByCode("jje"),
file = args.a == "y" and "Jje-" .. PAGENAME .. ".ogg" or args.a,
})
:done()
:done()
:done()
end
--[[if args.l then
html_li_ph
:tag( "ul" )
:tag( "li" )
:addClass( "ko-pron__note-vowel-length" )
:wikitext( 'Though still prescribed in Standard Korean, the great majority of speakers (in both Koreas) no longer distinguish vowel length.' )
:done()
:done()
:done()
end-]]
html_ul
:node( html_li_ipa_p )
:node( html_li_ipa )
:node( html_li_ph )
:done()
local html_table = mw.html.create( "table" )
:addClass( "jje-pron" )
:addClass( "mw-collapsible" )
:addClass( "mw-collapsed" )
:tag( "tr" )
:tag( "th" )
:attr( "colspan", 2 )
:wikitext( "Romanizations" )
:done()
:done()
:done()
for roman_index = 2, 4 do
html_table
:tag( "tr" )
:tag( "th" )
:wikitext( system_list[roman_index].display )
:done()
:tag( "td" )
:addClass( "IPA" )
:wikitext( concat(output_result[roman_index], system_list[roman_index].separator) )
:done()
:done()
:done()
end
return tostring(html_ul) .. tostring(html_table) .. require("Module:TemplateStyles")("Template:jje-IPA/style.css")
end
--[[function export.make_hanja(frame, scheme)
local params = {
[1] = { list = true },
["l"] = {},
}
local args = require("Module:parameters").process(frame:getParent().args, params)
local results = {
[1] = {},
[6] = {},
}
for _, text_param in ipairs(args[1]) do
for _, system_index in pairs({1, 6}) do
local romanised = export.romanise(text_param, system_index, args)
insert(results[system_index], romanised)
end
end
local html_ul = mw.html.create( "ul" )
:done()
local html_li_ipa = mw.html.create( "li" )
:wikitext( system_list[6].display )
:tag( "span" )
:addClass( "IPA" )
:wikitext( concat(results[6], system_list[6].separator) )
:done()
:done()
local html_li_ph = mw.html.create( "li" )
:addClass( "ko-pron__ph" )
:wikitext( system_list[1].display )
:tag( "span" )
:addClass( "Kore" )
:attr( "lang", "ko" )
:wikitext( "[" .. concat(results[1], system_list[1].separator) .. "]" )
:done()
:done()
if args.l then
html_li_ph
:tag( "ul" )
:tag( "li" )
:addClass( "ko-pron__note-vowel-length" )
:wikitext( 'Long vowel distinction only applies at the initial position. Most speakers no longer distinguish vowel length at any position.' )
:done()
:done()
:done()
end
html_ul
:node( html_li_ipa )
:node( html_li_ph )
:done()
return tostring(html_ul) .. require("Module:TemplateStyles")("Template:ko-IPA/style.css")
end]]
return export