Module:languages: difference between revisions

From Wiktionary, the free dictionary
Jump to navigation Jump to search
Browse history interactively
← Go to previous edit
Content deleted Content added
Track "nan-hok" (etym-only language), which is being converted to "nan-hbl" (full language).
Tags: Mobile edit Mobile web edit Advanced mobile edit
Further performance enhancements.
 
(40 intermediate revisions by 4 users not shown)
Line 1: Line 1:
local require = require

local m_str_utils = require("Module:string utilities")
local m_table = require("Module:table")
local mw = mw
local string = string
local table = table
local ustring = mw.ustring

local check_object = require("Module:utilities").check_object
local concat = table.concat
local decode_entities = m_str_utils.decode_entities
local decode_uri = m_str_utils.decode_uri
local gmatch = string.gmatch
local gsub = string.gsub
local insert = table.insert
local insert = table.insert
local ipairs = ipairs
local load_data = mw.loadData
local match = string.match
local next = next
local pairs = pairs
local remove = table.remove
local remove = table.remove
local remove_duplicates = m_table.removeDuplicates
local ugsub = mw.ustring.gsub
local select = select
local setmetatable = setmetatable
local shallowcopy = m_table.shallowcopy
local split = m_str_utils.split
local type = type
local ugsub = ustring.gsub
local ulen = m_str_utils.len
local umatch = ustring.match

-- Loaded as needed by findBestScript.
local Hans_chars
local Hant_chars


local export = {}
local export = {}
Line 121: Line 153:


local function normalize_code(code)
local function normalize_code(code)
return mw.loadData("Module:languages/data").aliases[code] or code
return load_data("Module:languages/data").aliases[code] or code
end

local checkObject = require("Module:utilities").check_object

-- Remove directional formatting characters from the start and end of the text.
local dir_char = "\226\128\170-\226\128\174\226\129\166-\226\129\169"
local function remove_directional_chars(text)
return (ugsub(text, "^[" .. dir_char .. "]*(.*)%f[%z" .. dir_char .. "][" .. dir_char .. "]*$", "%1"))
end
end


-- Convert risky characters to HTML entities, which minimizes interference once returned (e.g. for "sms:a", "<!-- -->" etc.).
-- Convert risky characters to HTML entities, which minimizes interference once returned (e.g. for "sms:a", "<!-- -->" etc.).
local function escape_risky_characters(text)
local function escape_risky_characters(text)
local make_entities = require("Module:utilities").make_entities
local encode_entities = require("Module:string/encode entities")
-- Spacing characters in isolation generally need to be escaped in order to be properly processed by the MediaWiki software.
-- Spacing characters in isolation generally need to be escaped in order to be properly processed by the MediaWiki software.
if not mw.ustring.match(text, "%S") then
if umatch(text, "^%s*$") then
return make_entities(text, text)
return encode_entities(text, text)
else
else
return make_entities(text, "!#%&*+/:;<=>?@[\\]_{|}")
return encode_entities(text, "!#%&*+/:;<=>?@[\\]_{|}")
end
end
end
end


-- Temporarily convert various formatting characters to PUA to prevent them from being disrupted by the substitution process.
local function make_language(code, data, useRequire)
local function conditionalRequire(modulename)
local function doTempSubstitutions(text, subbedChars, keepCarets, noTrim)
-- Clone so that we don't insert any extra patterns into the table in package.loaded. For some reason, using require seems to keep memory use down; probably because the table is always cloned.
if useRequire then
local patterns = shallowcopy(require("Module:languages/data/patterns"))
return require(modulename)
if keepCarets then
else
insert(patterns, "((\\+)%^)")
return mw.loadData(modulename)
insert(patterns, "((%^))")
end
end
end
-- Ensure any whitespace at the beginning and end is temp substituted, to prevent it from being accidentally trimmed. We only want to trim any final spaces added during the substitution process (e.g. by a module), which means we only do this during the first round of temp substitutions.
if not noTrim then
-- Temporarily convert various formatting characters to PUA to prevent them from being disrupted by the substitution process.
insert(patterns, "^([\128-\191\244]*(%s+))")
local function doTempSubstitutions(text, subbedChars, keepCarets, noTrim)
insert(patterns, "((%s+)[\128-\191\244]*)$")
-- Clone so that we don't insert any extra patterns into the table in package.loaded. For some reason, using require seems to keep memory use down; probably because the table is always cloned.
local patterns = require("Module:table").shallowcopy(require("Module:languages/data/patterns"))
if keepCarets then
insert(patterns, "((\\+)%^)")
insert(patterns, "((%^))")
end
-- Ensure any whitespace at the beginning and end is temp substituted, to prevent it from being accidentally trimmed. We only want to trim any final spaces added during the substitution process (e.g. by a module), which means we only do this during the first round of temp substitutions.
if not noTrim then
insert(patterns, "^([\128-\191\244]*(%s+))")
insert(patterns, "((%s+)[\128-\191\244]*)$")
end
-- Pre-substitution, of "[[" and "]]", which makes pattern matching more accurate.
text = text
:gsub("%f[%[]%[%[", "\1")
:gsub("%f[%]]%]%]", "\2")
local i, pe = #subbedChars, require("Module:string/pattern escape")
for j, pattern in ipairs(patterns) do
-- Patterns ending in \0 stand are for things like "[[" or "]]"), so the inserted PUA are treated as breaks between terms by modules that scrape info from pages.
local term_divider
pattern = pattern:gsub("%z$", function(divider)
term_divider = divider == "\0"
return ""
end)
text = text:gsub(pattern, function(...)
local m = {...}
local m1New = m[1]
for k = 2, #m do
local n = i + k - 1
subbedChars[n] = m[k]
local byte2 = math.floor(n / 4096) % 64 + (term_divider and 128 or 136)
local byte3 = math.floor(n / 64) % 64 + 128
local byte4 = n % 64 + 128
m1New = m1New:gsub(pe(m[k]), "\244" .. string.char(byte2) .. string.char(byte3) .. string.char(byte4), 1)
end
i = i + #m - 1
return m1New
end)
end
text = text
:gsub("\1", "%[%[")
:gsub("\2", "%]%]")
return text, subbedChars
end
end
-- Pre-substitution, of "[[" and "]]", which makes pattern matching more accurate.
text = text
-- Reinsert any formatting that was temporarily substituted.
:gsub("%f[%[]%[%[", "\1")
local function undoTempSubstitutions(text, subbedChars)
:gsub("%f[%]]%]%]", "\2")
local pe = require("Module:string/pattern escape")
for i = 1, #subbedChars do
local i, pe = #subbedChars, require("Module:string utilities").pattern_escape
for _, pattern in ipairs(patterns) do
local byte2 = math.floor(i / 4096) % 64 + 128
-- Patterns ending in \0 stand are for things like "[[" or "]]"), so the inserted PUA are treated as breaks between terms by modules that scrape info from pages.
local byte3 = math.floor(i / 64) % 64 + 128
local byte4 = i % 64 + 128
local term_divider
pattern = pattern:gsub("%z$", function(divider)
text = text:gsub("\244[" .. string.char(byte2) .. string.char(byte2+8) .. "]" .. string.char(byte3) .. string.char(byte4), pe(subbedChars[i]))
term_divider = divider == "\0"
end
return ""
text = text
end)
:gsub("\1", "%[%[")
:gsub("\2", "%]%]")
text = text:gsub(pattern, function(...)
local m = {...}
return text
local m1New = m[1]
for k = 2, #m do
local n = i + k - 1
subbedChars[n] = m[k]
local byte2 = math.floor(n / 4096) % 64 + (term_divider and 128 or 136)
local byte3 = math.floor(n / 64) % 64 + 128
local byte4 = n % 64 + 128
m1New = m1New:gsub(pe(m[k]), "\244" .. string.char(byte2) .. string.char(byte3) .. string.char(byte4), 1)
end
i = i + #m - 1
return m1New
end)
end
end
text = text
:gsub("\1", "%[%[")
-- Convert any HTML entities.
:gsub("\2", "%]%]")
local function noEntities(text)
return text, subbedChars
if text:match("&[^;]+;") then
end
return require("Module:utilities").get_entities(text)

else
-- Reinsert any formatting that was temporarily substituted.
return text
local function undoTempSubstitutions(text, subbedChars)
end
local pe = require("Module:string utilities").pattern_escape
for i = 1, #subbedChars do
local byte2 = math.floor(i / 4096) % 64 + 128
local byte3 = math.floor(i / 64) % 64 + 128
local byte4 = i % 64 + 128
text = text:gsub("\244[" .. string.char(byte2) .. string.char(byte2+8) .. "]" .. string.char(byte3) .. string.char(byte4), pe(subbedChars[i]))
end
end
text = text
:gsub("\1", "%[%[")
-- Check if the raw text is an unsupported title, and if so return that. Otherwise, remove HTML entities. We do the pre-conversion to avoid loading the unsupported title list unnecessarily.
:gsub("\2", "%]%]")
local function checkNoEntities(text)
return text
local textNoEnc = noEntities(text)
end
if textNoEnc ~= text and conditionalRequire("Module:links/data").unsupported_titles[text] then

return text
-- Check if the raw text is an unsupported title, and if so return that. Otherwise, remove HTML entities. We do the pre-conversion to avoid loading the unsupported title list unnecessarily.
else
local function checkNoEntities(self, text)
return textNoEnc
local textNoEnc = decode_entities(text)
end
if textNoEnc ~= text and self:loadData("Module:links/data").unsupported_titles[text] then
return text
else
return textNoEnc
end
end
end

-- If no script object is provided (or if it's invalid or None), get one.
-- If no script object is provided (or if it's invalid or None), get one.
local function checkScript(text, self, sc)
local function checkScript(text, self, sc)
if not checkObject("script", true, sc) or sc:getCode() == "None" then
if not check_object("script", true, sc) or sc:getCode() == "None" then
return self:findBestScript(text)
return self:findBestScript(text)
else
else
return sc
return sc
end
end
end
end

local function normalize(text, sc)
local function normalize(text, sc)
text = sc:fixDiscouragedSequences(text)
return sc:toFixedNFD(text)
text = sc:fixDiscouragedSequences(text)
return sc:toFixedNFD(text)
end

-- Split the text into sections, based on the presence of temporarily substituted formatting characters, then iterate over each one to apply substitutions. This avoids putting PUA characters through language-specific modules, which may be unequipped for them.
local function iterateSectionSubstitutions(text, subbedChars, keepCarets, self, sc, substitution_data, function_name)
local pe = require("Module:string utilities").pattern_escape
local fail, cats, sections = nil, {}
-- See [[Module:languages/data]].
if not text:find("\244") or self:loadData("Module:languages/data").contiguous_substitution[self._code] then
sections = {text}
else
sections = split(text, "\244[\128-\143][\128-\191]*", true)
end
end
for _, section in ipairs(sections) do
-- Don't bother processing empty strings or whitespace (which may also not be handled well by dedicated modules).
-- Split the text into sections, based on the presence of temporarily substituted formatting characters, then iterate over each one to apply substitutions. This avoids putting PUA characters through language-specific modules, which may be unequipped for them.
if section:gsub("%s+", "") ~= "" then
local function iterateSectionSubstitutions(text, subbedChars, keepCarets, self, sc, substitution_data, function_name)
local pe = require("Module:string/pattern escape")
local sub, sub_fail, sub_cats = require("Module:languages/doSubstitutions")(section, self, sc, substitution_data, function_name)
-- Second round of temporary substitutions, in case any formatting was added by the main substitution process. However, don't do this if the section contains formatting already (as it would have had to have been escaped to reach this stage, and therefore should be given as raw text).
local fail, cats, sections = nil, {}
if sub and subbedChars then
-- See [[Module:languages/data]].
local noSub
if not text:match("\244") or conditionalRequire("Module:languages/data").contiguous_substitution[self:getCode()] then
for _, pattern in ipairs(require("Module:languages/data/patterns")) do
sections = {text}
if section:match(pattern .. "%z?") then
else
noSub = true
sections = mw.text.split(text, "[􀀀-􏿽]")
end
for i, section in ipairs(sections) do
-- Don't bother processing empty strings or whitespace (which may also not be handled well by dedicated modules).
if section:gsub("%s", "") ~= "" then
local sub, sub_fail, sub_cats = require("Module:languages/doSubstitutions")(section, self, sc, substitution_data, function_name)
-- Second round of temporary substitutions, in case any formatting was added by the main substitution process. However, don't do this if the section contains formatting already (as it would have had to have been escaped to reach this stage, and therefore should be given as raw text).
if sub and subbedChars then
local noSub
for _, pattern in ipairs(require("Module:languages/data/patterns")) do
if section:match(pattern .. "%z?") then
noSub = true
end
end
if not noSub then
sub, subbedChars = doTempSubstitutions(sub, subbedChars, keepCarets, true)
end
end
end
end
if (not sub) or sub_fail then
if not noSub then
sub, subbedChars = doTempSubstitutions(sub, subbedChars, keepCarets, true)
text = sub
fail = sub_fail
cats = sub_cats or {}
break
end
end
end
text = sub and text:gsub(pe(section), pe(sub), 1) or text
if type(sub_cats) == "table" then
if (not sub) or sub_fail then
text = sub
for _, cat in ipairs(sub_cats) do
fail = sub_fail
insert(cats, cat)
cats = sub_cats or {}
end
break
end
text = sub and text:gsub(pe(section), pe(sub), 1) or text
if type(sub_cats) == "table" then
for _, cat in ipairs(sub_cats) do
insert(cats, cat)
end
end
end
end
end
end
-- Trim, unless there are only spacing characters, while ignoring any final formatting characters.
text = text and text
:gsub("^([\128-\191\244]*)%s+(%S)", "%1%2")
:gsub("(%S)%s+([\128-\191\244]*)$", "%1%2")
-- Remove duplicate categories.
if #cats > 1 then
cats = require("Module:table").removeDuplicates(cats)
end
return text, fail, cats, subbedChars
end
end

-- Trim, unless there are only spacing characters, while ignoring any final formatting characters.
-- Process carets (and any escapes). Default to simple removal, if no pattern/replacement is given.
text = text and text
local function processCarets(text, pattern, repl)
:gsub("^([\128-\191\244]*)%s+(%S)", "%1%2")
local rep
:gsub("(%S)%s+([\128-\191\244]*)$", "%1%2")
repeat

text, rep = text:gsub("\\\\(\\*^)", "\3%1")
-- Remove duplicate categories.
until rep == 0
if #cats > 1 then
cats = remove_duplicates(cats)
end

return text, fail, cats, subbedChars
end

-- Process carets (and any escapes). Default to simple removal, if no pattern/replacement is given.
local function processCarets(text, pattern, repl)
local rep
repeat
text, rep = text:gsub("\\\\(\\*^)", "\3%1")
until rep == 0
return text
:gsub("\\^", "\4")
:gsub(pattern or "%^", repl or "")
:gsub("\3", "\\")
:gsub("\4", "^")
end

-- Remove carets if they are used to capitalize parts of transliterations (unless they have been escaped).
local function removeCarets(text, sc)
if not sc:hasCapitalization() and sc:isTransliterated() and text:match("%^") then
return processCarets(text)
else
return text
return text
:gsub("\\^", "\4")
:gsub(pattern or "%^", repl or "")
:gsub("\3", "\\")
:gsub("\4", "^")
end
end
end

-- Remove carets if they are used to capitalize parts of transliterations (unless they have been escaped).
local function removeCarets(text, sc)
local Language = {}

if not sc:hasCapitalization() and sc:isTransliterated() and text:match("%^") then
function Language:loadData(modname)
return processCarets(text)
return (self._useRequire and require or mw.loadData)(modname)
else
end
return text

end
--[==[Returns the language code of the language. Example: {{code|lua|"fr"}} for French.]==]
function Language:getCode()
return self._code
end

--[==[Returns the canonical name of the language. This is the name used to represent that language on Wiktionary, and is guaranteed to be unique to that language alone. Example: {{code|lua|"French"}} for French.]==]
function Language:getCanonicalName()
local name = self._name
if name == nil then
name = self._rawData[1]
self._name = name
end
end
return name
end
local Language = {}

--[==[
--[==[Returns the language code of the language. Example: {{code|lua|"fr"}} for French.]==]
Return the display form of the language. The display form of a language, family or script is the form it takes when
function Language:getCode()
appearing as the <code><var>source</var></code> in categories such as <code>English terms derived from
return self._code
<var>source</var></code> or <code>English given names from <var>source</var></code>, and is also the displayed text
end
in {makeCategoryLink()} links. For full and etymology-only languages, this is the same as the canonical name, but
for families, it reads <code>"<var>name</var> languages"</code> (e.g. {"Indo-Iranian languages"}), and for scripts,
--[==[Returns the canonical name of the language. This is the name used to represent that language on Wiktionary, and is guaranteed to be unique to that language alone. Example: {{code|lua|"French"}} for French.]==]
it reads <code>"<var>name</var> script"</code> (e.g. {"Arabic script"}).
function Language:getCanonicalName()
]==]
return self._rawData[1]
function Language:getDisplayForm()
end
local form = self._displayForm
if form == nil then
--[==[Returns the display form of the language. The display form of a language, family or script is the form it takes when appearing as the ''SOURCE'' in categories such as <code>English terms derived from ''SOURCE''</code> or <code>English given names from ''SOURCE''</code>, and is also the displayed text in <code>:makeCategoryLink</code> links. For regular and etymology languages, this is the same as the canonical name, but for families, it reads "NAME languages" (e.g. {{code|lua|"Indo-Iranian languages"}}), and for scripts, it reads "NAME script" (e.g. {{code|lua|"Arabic script"}}).]==]
form = self:getCanonicalName()
function Language:getDisplayForm()
-- Add article and " substrate" to substrates that lack them.
if not self._displayForm then
if self:getFamilyCode() == "qfa-sub" then
local form = self:getCanonicalName()
if not (match(form, "^[Tt]he ") or match(form, "^[Aa] ")) then
-- Add article and " substrate" if a substrate that lacks them.
form = "a " .. form
if self:getFamilyCode() == "qfa-sub" then
end
if not (form:find("^[Tt]he ") or form:find("^[Aa] ")) then
if not match(form, "[Ss]ubstrate") then
form = "a " .. form
form = form .. " substrate"
end
if not form:find("[Ss]ubstrate") then
form = form .. " substrate"
end
end
end
self._displayForm = form
end
end
return self._displayForm
self._displayForm = form
end
end
return form
end
--[==[Returns a table of the "other names" that the language is known by, excluding the canonical name. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {{code|lua|{"Manx Gaelic", "Northern Manx", "Southern Manx"} }} for [[:Category:Manx language|Manx]]. If <code>onlyOtherNames</code> is given and is non-{{code|lua|nil}}, only names explicitly listed in the <code>otherNames</code> field are returned; otherwise, names listed under <code>otherNames</code>, <code>aliases</code> and <code>varieties</code> are combined together and returned. For example, for Manx, Manx Gaelic is listed as an alias, while Northern Manx and Southern Manx are listed as varieties. It should be noted that the <code>otherNames</code> field itself is deprecated, and entries listed there should eventually be moved to either <code>aliases</code> or <code>varieties</code>.]==]

function Language:getOtherNames(onlyOtherNames)
--[==[Returns a table of the "other names" that the language is known by, excluding the canonical name. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {{code|lua|{"Manx Gaelic", "Northern Manx", "Southern Manx"} }} for [[:Category:Manx language|Manx]]. If <code>onlyOtherNames</code> is given and is non-{{code|lua|nil}}, only names explicitly listed in the <code>otherNames</code> field are returned; otherwise, names listed under <code>otherNames</code>, <code>aliases</code> and <code>varieties</code> are combined together and returned. For example, for Manx, Manx Gaelic is listed as an alias, while Northern Manx and Southern Manx are listed as varieties. It should be noted that the <code>otherNames</code> field itself is deprecated, and entries listed there should eventually be moved to either <code>aliases</code> or <code>varieties</code>.]==]
if #self._stack == 1 then
function Language:getOtherNames(onlyOtherNames)
self:loadInExtraData()
if #self._stack == 1 then
end
self:loadInExtraData()
return require("Module:language-like").getOtherNames(self, onlyOtherNames)
end
end
return require("Module:language-like").getOtherNames(self, onlyOtherNames)
end
--[==[Returns a table of the aliases that the language is known by, excluding the canonical name. Aliases are synonyms for the language in question. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {{code|lua|{"High German", "New High German", "Deutsch"} }} for [[:Category:German language|German]].]==]

function Language:getAliases()
--[==[Returns a table of the aliases that the language is known by, excluding the canonical name. Aliases are synonyms for the language in question. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {{code|lua|{"High German", "New High German", "Deutsch"} }} for [[:Category:German language|German]].]==]
if #self._stack == 1 then
function Language:getAliases()
self:loadInExtraData()
if #self._stack == 1 then
end
self:loadInExtraData()
return self._rawData.aliases or (self._extraData and self._extraData.aliases) or {}
end
end
return self._rawData.aliases or (self._extraData and self._extraData.aliases) or {}
end
--[==[Returns a table of the known subvarieties of a given language, excluding subvarieties that have been given explicit etymology language codes. The names are not guaranteed to be unique, in that sometimes a given name refers to a subvariety of more than one language. Example: {{code|lua|{"Southern Aymara", "Central Aymara"} }} for [[:Category:Aymara language|Aymara]]. Note that the returned value can have nested tables in it, when a subvariety goes by more than one name. Example: {{code|lua|{"North Azerbaijani", "South Azerbaijani", {"Afshar", "Afshari", "Afshar Azerbaijani", "Afchar"}, {"Qashqa'i", "Qashqai", "Kashkay"}, "Sonqor"} }} for [[:Category:Azerbaijani language|Azerbaijani]]. Here, for example, Afshar, Afshari, Afshar Azerbaijani and Afchar all refer to the same subvariety, whose preferred name is Afshar (the one listed first). To avoid a return value with nested tables in it, specify a non-{{code|lua|nil}} value for the <code>flatten</code> parameter; in that case, the return value would be {{code|lua|{"North Azerbaijani", "South Azerbaijani", "Afshar", "Afshari", "Afshar Azerbaijani", "Afchar", "Qashqa'i", "Qashqai", "Kashkay", "Sonqor"} }}.]==]

function Language:getVarieties(flatten)
--[==[
if #self._stack == 1 then
Return a table of the known subvarieties of a given language, excluding subvarieties that have been given
self:loadInExtraData()
explicit etymology-only language codes. The names are not guaranteed to be unique, in that sometimes a given name
end
refers to a subvariety of more than one language. Example: {{code|lua|{"Southern Aymara", "Central Aymara"} }} for
return require("Module:language-like").getVarieties(self, flatten)
[[:Category:Aymara language|Aymara]]. Note that the returned value can have nested tables in it, when a subvariety
goes by more than one name. Example: {{code|lua|{"North Azerbaijani", "South Azerbaijani", {"Afshar", "Afshari",
"Afshar Azerbaijani", "Afchar"}, {"Qashqa'i", "Qashqai", "Kashkay"}, "Sonqor"} }} for
[[:Category:Azerbaijani language|Azerbaijani]]. Here, for example, Afshar, Afshari, Afshar Azerbaijani and Afchar
all refer to the same subvariety, whose preferred name is Afshar (the one listed first). To avoid a return value
with nested tables in it, specify a non-{{code|lua|nil}} value for the <code>flatten</code> parameter; in that case,
the return value would be {{code|lua|{"North Azerbaijani", "South Azerbaijani", "Afshar", "Afshari",
"Afshar Azerbaijani", "Afchar", "Qashqa'i", "Qashqai", "Kashkay", "Sonqor"} }}.
]==]
function Language:getVarieties(flatten)
if #self._stack == 1 then
self:loadInExtraData()
end
end
return require("Module:language-like").getVarieties(self, flatten)
end
--[==[Given a list of types as strings, returns true if the language has all of them.

--[==[Returns a table of types as a lookup table (with the types as keys).


The possible types are
The possible types are
* {language}: This is a language, either full or etymology-only.
* {language}: This is a language, either full or etymology-only.
* {full}: This is a "full" (not etymology-only) language, i.e. the union of {regular}, {reconstructed} and
* {full}: This is a "full" (not etymology-only) language, i.e. the union of {regular}, {reconstructed} and
{appendix-constructed}. Note that the types {full} and {etymology-only} also exist for families, so if you
{appendix-constructed}. Note that the types {full} and {etymology-only} also exist for families, so if you
want to check specifically for a full language and you have an object that might be a family, you should
want to check specifically for a full language and you have an object that might be a family, you should
use {{lua|hasType("language", "full")}} and not simply {{lua|hasType("full")}}.
use {{lua|hasType("language", "full")}} and not simply {{lua|hasType("full")}}.
* {etymology-only}: This is an etymology-only (not full) language, whose parent is another etymology-only
* {etymology-only}: This is an etymology-only (not full) language, whose parent is another etymology-only
language or a full language. Note that the types {full} and {etymology-only} also exist for
language or a full language. Note that the types {full} and {etymology-only} also exist for
families, so if you want to check specifically for an etymology-only language and you have an
families, so if you want to check specifically for an etymology-only language and you have an
object that might be a family, you should use {{lua|hasType("language", "etymology-only")}}
object that might be a family, you should use {{lua|hasType("language", "etymology-only")}}
and not simply {{lua|hasType("etymology-only")}}.
and not simply {{lua|hasType("etymology-only")}}.
* {regular}: This indicates a full language that is attested according to [[WT:CFI]] and therefore permitted
* {regular}: This indicates a full language that is attested according to [[WT:CFI]] and therefore permitted
in the main namespace. There may also be reconstructed terms for the language, which are placed in
in the main namespace. There may also be reconstructed terms for the language, which are placed in
the {Reconstruction} namespace and must be prefixed with * to indicate a reconstruction. Most full
the {Reconstruction} namespace and must be prefixed with * to indicate a reconstruction. Most full
languages are natural (not constructed) languages, but a few constructed languages (e.g. Esperanto
languages are natural (not constructed) languages, but a few constructed languages (e.g. Esperanto
and Volapük, among others) are also allowed in the mainspace and considered regular languages.
and Volapük, among others) are also allowed in the mainspace and considered regular languages.
* {reconstructed}: This language is not attested according to [[WT:CFI]], and therefore is allowed only in the
* {reconstructed}: This language is not attested according to [[WT:CFI]], and therefore is allowed only in the
{Reconstruction} namespace. All terms in this language are reconstructed, and must be prefixed
{Reconstruction} namespace. All terms in this language are reconstructed, and must be prefixed
with *. Languages such as Proto-Indo-European and Proto-Germanic are in this category.
with *. Languages such as Proto-Indo-European and Proto-Germanic are in this category.
* {appendix-constructed}: This language is attested but does not meet the additional requirements set out for
* {appendix-constructed}: This language is attested but does not meet the additional requirements set out for
constructed languages ([[WT:CFI#Constructed languages]]). Its entries must therefore
constructed languages ([[WT:CFI#Constructed languages]]). Its entries must therefore
be in the Appendix namespace, but they are not reconstructed and therefore should
be in the Appendix namespace, but they are not reconstructed and therefore should
not have * prefixed in links.
not have * prefixed in links.
]==]
]==]
function Language:hasType(...)
function Language:getTypes()
if not self._type then
local types = self._types
if types == nil then
self._type = {language = true}
types = {language = true}
if self:getNonEtymologicalCode() == self:getCode() then
self._type.full = true
if self:getFullCode() == self._code then
types.full = true
else
else
self._type["etymology-only"] = true
types["etymology-only"] = true
end
for _, type in ipairs(mw.text.split(self._rawData.type, "%s*,%s*")) do
self._type[type] = true
end
end
end
for _, type in ipairs{...} do
for t in gmatch(self._rawData.type, "[^,]+") do
types[t] = true
if not self._type[type] then
return false
end
end
end
self._types = types
return true
end
end
return types
end
--[==[Returns a table containing <code>WikimediaLanguage</code> objects (see [[Module:wikimedia languages]]), which represent languages and their codes as they are used in Wikimedia projects for interwiki linking and such. More than one object may be returned, as a single Wiktionary language may correspond to multiple Wikimedia languages. For example, Wiktionary's single code <code>sh</code> (Serbo-Croatian) maps to four Wikimedia codes: <code>sh</code> (Serbo-Croatian), <code>bs</code> (Bosnian), <code>hr</code> (Croatian) and <code>sr</code> (Serbian).

The code for the Wikimedia language is retrieved from the <code>wikimedia_codes</code> property in the data modules. If that property is not present, the code of the current language is used. If none of the available codes is actually a valid Wikimedia code, an empty table is returned.]==]
--[==[Given a list of types as strings, returns true if the language has all of them.]==]
function Language:getWikimediaLanguages()
function Language:hasType(...)
if not self._wikimediaLanguageObjects then
local args, types = {...}, self:getTypes()
local m_wikimedia_languages = require("Module:wikimedia languages")
for i = 1, #args do
self._wikimediaLanguageObjects = {}
if not types[args[i]] then
local wikimedia_codes = self:getWikimediaLanguageCodes()
return false
for _, wlangcode in ipairs(wikimedia_codes) do
insert(self._wikimediaLanguageObjects, m_wikimedia_languages.getByCode(wlangcode))
end
end
end
return self._wikimediaLanguageObjects
end
end
return true
end
function Language:getWikimediaLanguageCodes()

if not self._wikimediaLanguageCodes then
--[==[Returns a table containing <code>WikimediaLanguage</code> objects (see [[Module:wikimedia languages]]), which represent languages and their codes as they are used in Wikimedia projects for interwiki linking and such. More than one object may be returned, as a single Wiktionary language may correspond to multiple Wikimedia languages. For example, Wiktionary's single code <code>sh</code> (Serbo-Croatian) maps to four Wikimedia codes: <code>sh</code> (Serbo-Croatian), <code>bs</code> (Bosnian), <code>hr</code> (Croatian) and <code>sr</code> (Serbian).
self._wikimediaLanguageCodes = self._rawData.wikimedia_codes or {self:getCode()}
The code for the Wikimedia language is retrieved from the <code>wikimedia_codes</code> property in the data modules. If that property is not present, the code of the current language is used. If none of the available codes is actually a valid Wikimedia code, an empty table is returned.]==]
function Language:getWikimediaLanguages()
local wm_langs = self._wikimediaLanguageObjects
if wm_langs == nil then
local get_wm_lang = require("Module:wikimedia languages").getByCode
local codes = self:getWikimediaLanguageCodes()
wm_langs = {}
for i = 1, #codes do
wm_langs[i] = get_wm_lang(codes[i])
end
end
self._wikimediaLanguageObjects = wm_langs
return self._wikimediaLanguageCodes
end
end
return wm_langs
end
--[==[

Returns the name of the Wikipedia article for the language. If the property {wikipedia_article} is present in the
function Language:getWikimediaLanguageCodes()
data module it will be used first, otherwise a sitelink will be generated from {:getWikidataItem} (if set). Otherwise
local wm_langs = self._wikimediaLanguageCodes
{:getCategoryName} is used as fallback, unless `noCategoryFallback` is specified, in which case {nil} is returned.]==]
if wm_langs == nil then
function Language:getWikipediaArticle(noCategoryFallback)
wm_langs = self._rawData.wikimedia_codes
if self._wikipedia_article == nil then
wm_langs = wm_langs and split(wm_langs, ",", true, true) or {self._code}
if self._rawData.wikipedia_article then
self._wikimediaLanguageCodes = wm_langs
self._wikipedia_article = self._rawData.wikipedia_article
end
elseif self:getWikidataItem() and mw.wikibase then
return wm_langs
self._wikipedia_article = mw.wikibase.sitelink(self:getWikidataItem(), 'enwiki')
end
end

if not self._wikipedia_article then
--[==[
self._wikipedia_article = false
Returns the name of the Wikipedia article for the language. `project` specifies the language and project to retrieve
end
the article from, defaulting to {"enwiki"} for the English Wikipedia. Normally if specified it should be the project
code for a specific-language Wikipedia e.g. "zhwiki" for the Chinese Wikipedia, but it can be any project, including
non-Wikipedia ones. If the project is the English Wikipedia and the property {wikipedia_article} is present in the data
module it will be used first. In all other cases, a sitelink will be generated from {:getWikidataItem} (if set). The
resulting value (or lack of value) is cached so that subsequent calls are fast. If no value could be determined, and
`noCategoryFallback` is {false}, {:getCategoryName} is used as fallback; otherwise, {nil} is returned. Note that if
`noCategoryFallback` is {nil} or omitted, it defaults to {false} if the project is the English Wikipedia, otherwise
to {true}. In other words, under normal circumstances, if the English Wikipedia article couldn't be retrieved, the
return value will fall back to a link to the language's category, but this won't normally happen for any other project.
]==]
function Language:getWikipediaArticle(noCategoryFallback, project)
project = project or "enwiki"
local cached_value
if project == "enwiki" then
cached_value = self._wikipedia_article
if cached_value == nil then
cached_value = self._rawData.wikipedia_article
end
end
else
if not self._wikipedia_article and not noCategoryFallback then
-- If the project isn't enwiki, default to no category fallback, but this can be overridden by specifying the
return self:getCategoryName():gsub("Creole language", "Creole")
-- value `false`.
if noCategoryFallback == nil then
noCategoryFallback = true
end
end
return self._wikipedia_article or nil
if self._non_en_wikipedia_articles == nil then
self._non_en_wikipedia_articles = {}
end
cached_value = self._non_en_wikipedia_articles[project]
end
end
if cached_value == nil then -- not false
if self:getWikidataItem() and mw.wikibase then
function Language:makeWikipediaLink()
cached_value = mw.wikibase.sitelink(self:getWikidataItem(), project)
return "[[w:" .. self:getWikipediaArticle() .. "|" .. self:getCanonicalName() .. "]]"
end
if not cached_value then
cached_value = false
end
end
end
-- Now cache the determined value.
if project == "enwiki" then
--[==[Returns the Wikidata item id for the language or <code>nil</code>. This corresponds to the the second field in the data modules.]==]
self._wikipedia_article = cached_value
function Language:getWikidataItem()
else
if not self._WikidataItem then
self._non_en_wikipedia_articles[project] = cached_value
local item = self._rawData[2]
end
if type(item) == "number" then
if not cached_value and not noCategoryFallback then
self._WikidataItem = "Q" .. item
return self:getCategoryName():gsub("Creole language", "Creole")
else
end
self._WikidataItem = item
return cached_value or nil
end

function Language:makeWikipediaLink()
return "[[w:" .. self:getWikipediaArticle() .. "|" .. self:getCanonicalName() .. "]]"
end

--[==[Returns the Wikidata item id for the language or <code>nil</code>. This corresponds to the the second field in the data modules.]==]
function Language:getWikidataItem()
local item = self._WikidataItem
if item == nil then
item = self._rawData[2]
-- If the value is nil, it's cached as false.
item = item ~= nil and (type(item) == "number" and "Q" .. item or item) or false
self._WikidataItem = item
end
return item or nil
end

--[==[Returns a table of <code>Script</code> objects for all scripts that the language is written in. See [[Module:scripts]].]==]
function Language:getScripts()
local scripts = self._scriptObjects
if scripts == nil then
local codes = self:getScriptCodes()
if codes[1] == "All" then
scripts = self:loadData("Module:scripts/data")
else
local get_script = require("Module:scripts").getByCode
scripts = {}
for i = 1, #codes do
scripts[i] = get_script(codes[i], nil, nil, self._useRequire)
end
end
end
end
self._scriptObjects = scripts
return self._WikidataItem
end
end
return scripts
end
--[==[Returns a table of <code>Script</code> objects for all scripts that the language is written in. See [[Module:scripts]].]==]

function Language:getScripts()
--[==[Returns the table of script codes in the language's data file.]==]
if not self._scriptObjects then
function Language:getScriptCodes()
self._scriptObjects = {}
local scripts = self._scriptCodes
if self:getScriptCodes()[1] == "All" then
if scripts == nil then
self._scriptObjects = conditionalRequire("Module:scripts/data")
scripts = self._rawData[4]
else
if scripts then
for _, sc in ipairs(self:getScriptCodes()) do
local codes, n = {}, 0
insert(self._scriptObjects, require("Module:scripts").getByCode(sc, nil, nil, useRequire))
for code in gmatch(scripts, "[^,]+") do
n = n + 1
-- Special handling of "Hants", which represents "Hani", "Hant" and "Hans" collectively.
if code == "Hants" then
codes[n] = "Hani"
codes[n + 1] = "Hant"
codes[n + 2] = "Hans"
n = n + 2
else
codes[n] = code
end
end
end
end
scripts = codes
else
scripts = {"None"}
end
end
self._scriptCodes = scripts
return self._scriptObjects
end
end
return scripts
end

--[==[Given some text, this function iterates through the scripts of a given language and tries to find the script that best matches the text. It returns a {{code|lua|Script}} object representing the script. If no match is found at all, it returns the {{code|lua|None}} script object.]==]
function Language:findBestScript(text, forceDetect)
local useRequire = self._useRequire
if not text or text == "" or text == "-" then
--[==[Returns the table of script codes in the language's data file.]==]
return require("Module:scripts").getByCode("None", nil, nil, useRequire)
function Language:getScriptCodes()
if not self._scriptCodes then
self._scriptCodes = self._rawData[4] or {"None"}
end
return self._scriptCodes
end
end
-- Differs from table returned by getScriptCodes, as Hants is not normalized into its constituents.
--[==[Given some text, this function iterates through the scripts of a given language and tries to find the script that best matches the text. It returns a {{code|lua|Script}} object representing the script. If no match is found at all, it returns the {{code|lua|None}} script object.]==]
local codes = self._bestScriptCodes
function Language:findBestScript(text, forceDetect)
if (not text) or text == "" or text == "-" then
if codes == nil then
codes = self._rawData[4]
return require("Module:scripts").getByCode("None", nil, nil, useRequire)
codes = codes and split(codes, ",", true, true) or {"None"}
end
self._bestScriptCodes = codes
if table.concat(self:getScriptCodes()) == "All" then
return require("Module:scripts").findBestScriptWithoutLang(text)
end
local scripts = self:getScripts()
if not scripts[2] and not forceDetect then
-- Necessary, because Hani covers the entire Han range (while the Hant & Hans lists don't list shared characters).
if scripts[1]:getCode():match("^Han") and require("Module:scripts").getByCode("Hani", nil, nil, useRequire):countCharacters(text) > 0 then
return scripts[1]
elseif scripts[1]:countCharacters(text) > 0 then
return scripts[1]
else
return require("Module:scripts").getByCode("None", nil, nil, useRequire)
end
end
return require("Module:languages/findBestScript")(export, self, text, scripts, forceDetect, useRequire)
end
end
local first_sc = codes[1]
--[==[Returns a <code>Family</code> object for the language family that the language belongs to. See [[Module:families]].]==]
function Language:getFamily()
if self._familyObject == nil then
if first_sc == "All" then
return require("Module:scripts").findBestScriptWithoutLang(text)
local familyCode = self:getFamilyCode()
if familyCode then
self._familyObject = require("Module:families").getByCode(familyCode, useRequire)
-- Still memoize a nil result.
else
self._familyObject = false
end
end
return self._familyObject or nil
end
end
local get_script = require("Module:scripts").getByCode
--[==[Returns the family code in the language's data file.]==]
local codes_len = #codes
function Language:getFamilyCode()
if not self._familyCode then
if not (forceDetect or first_sc == "Hants" or codes_len > 1) then
self._familyCode = self._rawData[3]
first_sc = get_script(first_sc, nil, nil, useRequire)
end
local charset = first_sc.characters
return self._familyCode
return charset and umatch(text, "[" .. charset .. "]") and first_sc or
get_script("None", nil, nil, useRequire)
end
end
-- Remove all formatting characters.
function Language:getFamilyName()
text = require("Module:utilities").get_plaintext(text)
if self._familyName == nil then
local family = self:getFamily()
-- Remove all spaces and any ASCII punctuation. Some non-ASCII punctuation is script-specific, so can't be removed.
if family then
text = ugsub(text, "[%s!\"#%%&'()*,%-./:;?@[\\%]_{}]+", "")
self._familyName = family:getCanonicalName()
if #text == 0 then
else
return get_script("None", nil, nil, useRequire)
self._familyName = false
end
end
return self._familyName or nil
end
end
-- Try to match every script against the text,
--[==[Check whether the language belongs to `family` (which can be a family code or object). A list of objects can be given in place of `family`; in that case, return true if the language belongs to any of the specified families. Note that some languages (in particular, certain creoles) can have multiple immediate ancestors potentially belonging to different families; in that case, return true if the language belongs to any of the specified families.]==]
-- and return the one with the most matching characters.
function Language:inFamily(...)
local bestcount, bestscript, length = 0
--checkObject("family", nil, ...)
for _, family in ipairs{...} do
for i = 1, codes_len do
local sc = codes[i]
if type(family) == "table" then
-- Special case for "Hants", which is a special code that represents whichever of "Hant" or "Hans" best matches, or "Hani" if they match equally. This avoids having to list all three. In addition, "Hants" will be treated as the best match if there is at least one matching character, under the assumption that a Han script is desirable in terms that contain a mix of Han and other scripts (not counting those which use Jpan or Kore).
family = family:getCode()
if sc == "Hants" then
local Hani = get_script("Hani", nil, nil, useRequire)
if not Hant_chars then
Hant_chars = self:loadData("Module:zh/data/ts")
Hans_chars = self:loadData("Module:zh/data/st")
end
end
local self_family_code = self:getFamilyCode()
local t, s, found = 0, 0
-- This is faster than using mw.ustring.gmatch directly.
if not self_family_code then
for ch in gmatch(ugsub(text, "[" .. Hani.characters .. "]", "\255%0"), "\255(.[\128-\191]*)") do
return false
found = true
elseif self_family_code == family then
if Hant_chars[ch] then
return true
end
t = t + 1
if Hans_chars[ch] then
local self_family = self:getFamily()
s = s + 1
if self_family:inFamily(family) then
return true
-- If the family isn't a real family (e.g. creoles) check any ancestors.
elseif self_family:getFamilyCode() == "qfa-not" then
local ancestors = self:getAncestors()
for _, ancestor in ipairs(ancestors) do
if ancestor:inFamily(family) then
return true
end
end
elseif Hans_chars[ch] then
s = s + 1
else
t, s = t + 1, s + 1
end
end
end
end
end
if found then
return false
if t == s then
end
return Hani
end
function Language:getParent()
return get_script(t > s and "Hant" or "Hans", nil, nil, useRequire)
if self._parentObject == nil then
end
local parentCode = self:getParentCode()
else
if parentCode then
self._parentObject = export.getByCode(parentCode, nil, true, true, useRequire)
sc = get_script(sc, nil, nil, useRequire)
else
if not length then
self._parentObject = false
length = ulen(text)
end
-- Count characters by removing everything in the script's charset and comparing to the original length.
local charset = sc.characters
local count = charset and length - ulen(ugsub(text, "[" .. charset .. "]+", "")) or 0
if count >= length then
return sc
elseif count > bestcount then
bestcount = count
bestscript = sc
end
end
end
end
return self._parentObject or nil
end
end
-- Return best matching script, or otherwise None.
function Language:getParentCode()
return bestscript or get_script("None", nil, nil, useRequire)
if not self._parentCode then
end
self._parentCode = self._rawData[5]

end
--[==[Returns a <code>Family</code> object for the language family that the language belongs to. See [[Module:families]].]==]
return self._parentCode
function Language:getFamily()
local family = self._familyObject
if family == nil then
family = self:getFamilyCode()
-- If the value is nil, it's cached as false.
family = family and require("Module:families").getByCode(family, self._useRequire) or false
self._familyObject = family
end
end
return family or nil
end
function Language:getParentName()

if self._parentName == nil then
--[==[Returns the family code in the language's data file.]==]
local parent = self:getParent()
function Language:getFamilyCode()
if parent then
local family = self._familyCode
self._parentName = parent:getCanonicalName()
if family == nil then
else
-- If the value is nil, it's cached as false.
self._parentName = false
family = self._rawData[3] or false
end
self._familyCode = family
end
return self._parentName or nil
end
end
return family or nil
end
function Language:getParentChain()

if not self._parentChain then
function Language:getFamilyName()
self._parentChain = {}
local parent = self:getParent()
local family = self._familyName
if family == nil then
while parent do
family = self:getFamily()
insert(self._parentChain, parent)
-- If the value is nil, it's cached as false.
parent = parent:getParent()
family = family and family:getCanonicalName() or false
end
self._familyName = family
end
return family or nil
end

--[==[Check whether the language belongs to `family` (which can be a family code or object). A list of objects can be given in place of `family`; in that case, return true if the language belongs to any of the specified families. Note that some languages (in particular, certain creoles) can have multiple immediate ancestors potentially belonging to different families; in that case, return true if the language belongs to any of the specified families.]==]
function Language:inFamily(...)
--check_object("family", nil, ...)
for _, family in ipairs{...} do
if type(family) == "table" then
family = family:getCode()
end
end
local self_family_code = self:getFamilyCode()
return self._parentChain
if not self_family_code then
end
return false
elseif self_family_code == family then
function Language:hasParent(...)
return true
--checkObject("language", nil, ...)
end
for _, otherlang in ipairs{...} do
local self_family = self:getFamily()
for _, parent in ipairs(self:getParentChain()) do
if type(otherlang) == "string" then
if self_family:inFamily(family) then
if otherlang == parent:getCode() then return true end
return true
-- If the family isn't a real family (e.g. creoles) check any ancestors.
else
elseif self_family:getFamilyCode() == "qfa-not" then
if otherlang:getCode() == parent:getCode() then return true end
local ancestors = self:getAncestors()
for _, ancestor in ipairs(ancestors) do
if ancestor:inFamily(family) then
return true
end
end
end
end
end
end
return false
end
end
return false
end
--[==[If the language is an etymology language, this iterates through parents until a regular language or family is found, and the corresponding object is returned. If the language is a regular language, then it simply returns the language.]==]

function Language:getNonEtymological()
function Language:getParent()
if not self._nonEtymologicalObject then
local nonEtymologicalCode = self:getNonEtymologicalCode()
local parent = self._parentObject
if nonEtymologicalCode ~= self:getCode() then
if parent == nil then
parent = self:getParentCode()
self._nonEtymologicalObject = export.getByCode(nonEtymologicalCode, nil, nil, nil, useRequire)
-- If the value is nil, it's cached as false.
else
parent = parent and export.getByCode(parent, nil, true, true, self._useRequire) or false
self._nonEtymologicalObject = self
self._parentObject = parent
end
end
return self._nonEtymologicalObject
end
end
return parent or nil
end
function Language:getNonEtymologicalCode()

return self._nonEtymologicalCode or self:getCode()
function Language:getParentCode()
local parent = self._parentCode
if parent == nil then
-- If the value is nil, it's cached as false.
parent = self._rawData[5] or false
self._parentCode = parent
end
end
return parent or nil
end
function Language:getNonEtymologicalName()

if self._nonEtymologicalName == nil then
function Language:getParentName()
local nonEtymological = self:getNonEtymological()
local parent = self._parentName
if nonEtymological then
if parent == nil then
self._nonEtymologicalName = nonEtymological:getCanonicalName()
parent = self:getParent()
else
-- If the value is nil, it's cached as false.
self._nonEtymologicalName = false
parent = parent and parent:getCanonicalName() or false
end
self._parentName = parent
end
return parent or nil
end

function Language:getParentChain()
local chain = self._parentChain
if chain == nil then
chain = {}
local parent, n = self:getParent(), 0
while parent do
n = n + 1
chain[n] = parent
parent = parent:getParent()
end
end
return self._nonEtymologicalName or nil
self._parentChain = chain
end
end
return chain
end
--[==[Returns a table of <code class="nf">Language</code> objects for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.]==]

function Language:getAncestors()
function Language:hasParent(...)
if not self._ancestorObjects then
--check_object("language", nil, ...)
self._ancestorObjects = {}
for _, otherlang in ipairs{...} do
local ancestors = require("Module:table").shallowcopy(self:getAncestorCodes())
for _, parent in ipairs(self:getParentChain()) do
if #ancestors > 0 then
if type(otherlang) == "string" then
for _, ancestor in ipairs(ancestors) do
if otherlang == parent:getCode() then return true end
insert(self._ancestorObjects, export.getByCode(ancestor, nil, true, nil, useRequire))
end
else
else
if otherlang:getCode() == parent:getCode() then return true end
local fam = self:getFamily()
local protoLang = fam and fam:getProtoLanguage() or nil
-- For the cases where the current language is the proto-language
-- of its family, or an etymology language that is ancestral to that
-- proto-language, we need to step up a level higher right from the
-- start.
if protoLang and (
protoLang:getCode() == self:getCode() or
(self:hasType("etymology-only") and protoLang:hasAncestor(self))
) then
fam = fam:getFamily()
protoLang = fam and fam:getProtoLanguage() or nil
end
while not protoLang and not (not fam or fam:getCode() == "qfa-not") do
fam = fam:getFamily()
protoLang = fam and fam:getProtoLanguage() or nil
end
insert(self._ancestorObjects, protoLang)
end
end
end
end
return self._ancestorObjects
end
end
return false
end
do

-- Avoid a language being its own ancestor via class inheritance. We only need to check for this if the language has inherited an ancestor table from its parent, because we never want to drop ancestors that have been explicitly set in the data.
--[==[
-- Recursively iterate over ancestors until we either find self or run out. If self is found, return true.
If the language is etymology-only, this iterates through parents until a full language or family is found, and the
local function check_ancestor(self, lang)
corresponding object is returned. If the language is a full language, then it simply returns itself.
local codes = lang:getAncestorCodes()
]==]
if not codes then
function Language:getFull()
return nil
local full = self._fullObject
if full == nil then
full = self:getFullCode()
full = full == self._code and self or
export.getByCode(full, nil, nil, nil, self._useRequire)
self._fullObject = full
end
return full
end

--[==[
If the language is an etymology-only language, this iterates through parents until a full language or family is
found, and the corresponding code is returned. If the language is a full language, then it simply returns the
language code.
]==]
function Language:getFullCode()
return self._fullCode or self._code
end

--[==[
If the language is an etymology-only language, this iterates through parents until a full language or family is
found, and the corresponding canonical name is returned. If the language is a full language, then it simply returns
the canonical name of the language.
]==]
function Language:getFullName()
local full = self._fullName
if full == nil then
full = self:getFull():getCanonicalName()
self._fullName = full
end
return full
end

--[==[Returns a table of <code class="nf">Language</code> objects for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.]==]
function Language:getAncestors()
if not self._ancestorObjects then
self._ancestorObjects = {}
local ancestors = shallowcopy(self:getAncestorCodes())
if #ancestors > 0 then
for _, ancestor in ipairs(ancestors) do
insert(self._ancestorObjects, export.getByCode(ancestor, nil, true, nil, self._useRequire))
end
end
else
for i = 1, #codes do
local code = codes[i]
local fam = self:getFamily()
local protoLang = fam and fam:getProtoLanguage() or nil
if code == self:getCode() then
-- For the cases where the current language is the proto-language
return true
-- of its family, or an etymology-only language that is ancestral to that
end
-- proto-language, we need to step up a level higher right from the
local anc = export.getByCode(code, nil, true, nil, useRequire)
-- start.
if check_ancestor(self, anc) then
if protoLang and (
return true
protoLang:getCode() == self._code or
end
(self:hasType("etymology-only") and protoLang:hasAncestor(self))
end
end
--[==[Returns a table of <code class="nf">Language</code> codes for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.]==]
function Language:getAncestorCodes()
if self._ancestorCodes then
return self._ancestorCodes
elseif not self._rawData.ancestors then
self._ancestorCodes = {}
return self._ancestorCodes
end
local codes = self._rawData.ancestors
self._ancestorCodes = codes
if (
#codes == 0 or
#self._stack == 1 or
self._stack[#self._stack].ancestors
) then
) then
fam = fam:getFamily()
return codes
protoLang = fam and fam:getProtoLanguage() or nil
end
end
while not protoLang and not (not fam or fam:getCode() == "qfa-not") do
local i, code = 1
while i <= #codes do
fam = fam:getFamily()
protoLang = fam and fam:getProtoLanguage() or nil
code = codes[i]
if check_ancestor(self, self) then
remove(codes, i)
else
i = i + 1
end
end
end
insert(self._ancestorObjects, protoLang)
return codes
end
end
end
end
return self._ancestorObjects
end
--[==[Given a list of language objects or codes, returns true if at least one of them is an ancestor. This includes any etymology-only children of that ancestor. If the language's ancestor(s) are etymology-only languages, it will also return true for those language parent(s) (e.g. if Vulgar Latin is the ancestor, it will also return true for its parent, Latin). However, a parent is excluded from this if the ancestor is also ancestral to that parent (e.g. if Classical Persian is the ancestor, Persian would return false, because Classical Persian is also ancestral to Persian).]==]

function Language:hasAncestor(...)
do
--checkObject("language", nil, ...)
-- Avoid a language being its own ancestor via class inheritance. We only need to check for this if the language has inherited an ancestor table from its parent, because we never want to drop ancestors that have been explicitly set in the data.
-- Recursively iterate over ancestors until we either find self or run out. If self is found, return true.
local function iterateOverAncestorTree(node, func, parent_check)
local function check_ancestor(self, lang)
local ancestors = node:getAncestors()
local ancestorsParents = {}
local codes = lang:getAncestorCodes()
if not codes then
for _, ancestor in ipairs(ancestors) do
return nil
local ret = func(ancestor) or iterateOverAncestorTree(ancestor, func, parent_check)
if ret then return ret end
end
-- Check the parents of any ancestors. We don't do this if checking the parents of the other language, so that we exclude any etymology-only children of those parents that are not directly related (e.g. if the ancestor is Vulgar Latin and we are checking New Latin, we want it to return false because they are on different ancestral branches. As such, if we're already checking the parent of New Latin (Latin) we don't want to compare it to the parent of the ancestor (Latin), as this would be a false positive; it should be one or the other).
if not parent_check then
return nil
end
for _, ancestor in ipairs(ancestors) do
local ancestorParents = ancestor:getParentChain()
for _, ancestorParent in ipairs(ancestorParents) do
if ancestorParent:getCode() == self:getCode() or ancestorParent:hasAncestor(ancestor) then
break
else
insert(ancestorsParents, ancestorParent)
end
end
end
for _, ancestorParent in ipairs(ancestorsParents) do
local ret = func(ancestorParent)
if ret then return ret end
end
end
end
for i = 1, #codes do
local code = codes[i]
local function do_iteration(otherlang, parent_check)
-- otherlang can't be self
if code == self._code then
return true
if (type(otherlang) == "string" and otherlang or otherlang:getCode()) == self:getCode() then
return false
end
end
local anc = export.getByCode(code, nil, true, nil, self._useRequire)
repeat
if check_ancestor(self, anc) then
if iterateOverAncestorTree(
self,
function(ancestor)
return ancestor:getCode() == (type(otherlang) == "string" and otherlang or otherlang:getCode())
end,
parent_check
) then
return true
elseif type(otherlang) == "string" then
otherlang = export.getByCode(otherlang, nil, true, nil, useRequire)
end
otherlang = otherlang:getParent()
parent_check = false
until not otherlang
end
local parent_check = true
for _, otherlang in ipairs{...} do
local ret = do_iteration(otherlang, parent_check)
if ret then
return true
return true
end
end
end
end
return false
end
end

--[==[Returns a table of <code class="nf">Language</code> codes for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.]==]
function Language:getAncestorChain()
function Language:getAncestorCodes()
if not self._ancestorChain then
self._ancestorChain = {}
if self._ancestorCodes then
local step = self
return self._ancestorCodes
end
while true do
local ancestors = step:getAncestors()
local codes = self._rawData.ancestors
if not codes then
step = #ancestors == 1 and ancestors[1] or nil
codes = {}
if not step then break end
insert(self._ancestorChain, 1, step)
self._ancestorCodes = codes
return codes
end
codes = split(codes, ",", true, true)
self._ancestorCodes = codes
if (
#codes == 0 or
#self._stack == 1 or
self._stack[#self._stack].ancestors
) then
return codes
end
local i, code = 1
while i <= #codes do
code = codes[i]
if check_ancestor(self, self) then
remove(codes, i)
else
i = i + 1
end
end
end
end
return self._ancestorChain
return codes
end
end
end

local function fetch_descendants(self, format)
--[==[Given a list of language objects or codes, returns true if at least one of them is an ancestor. This includes any etymology-only children of that ancestor. If the language's ancestor(s) are etymology-only languages, it will also return true for those language parent(s) (e.g. if Vulgar Latin is the ancestor, it will also return true for its parent, Latin). However, a parent is excluded from this if the ancestor is also ancestral to that parent (e.g. if Classical Persian is the ancestor, Persian would return false, because Classical Persian is also ancestral to Persian).]==]
local languages = require("Module:languages/code to canonical name")
function Language:hasAncestor(...)
local etymology_languages = require("Module:etymology languages/code to canonical name")
--check_object("language", nil, ...)
local families = require("Module:families/code to canonical name")

local descendants = {}
local function iterateOverAncestorTree(node, func, parent_check)
local family = self:getFamily()
local ancestors = node:getAncestors()
-- Iterate over all three datasets.
local ancestorsParents = {}
for _, data in ipairs{languages, etymology_languages, families} do
for code in pairs(data) do
for _, ancestor in ipairs(ancestors) do
local ret = func(ancestor) or iterateOverAncestorTree(ancestor, func, parent_check)
local lang = export.getByCode(code, nil, true, true, useRequire)
if ret then return ret end
-- Test for a descendant. Earlier tests weed out most candidates, while the more intensive tests are only used sparingly.
if (
end
-- Check the parents of any ancestors. We don't do this if checking the parents of the other language, so that we exclude any etymology-only children of those parents that are not directly related (e.g. if the ancestor is Vulgar Latin and we are checking New Latin, we want it to return false because they are on different ancestral branches. As such, if we're already checking the parent of New Latin (Latin) we don't want to compare it to the parent of the ancestor (Latin), as this would be a false positive; it should be one or the other).
code ~= self:getCode() and -- Not self.
if not parent_check then
lang:inFamily(family) and -- In the same family.
(
return nil
end
family:getProtoLanguageCode() == self:getCode() or -- Self is the protolanguage.
for _, ancestor in ipairs(ancestors) do
self:hasDescendant(lang) or -- Full hasDescendant check.
local ancestorParents = ancestor:getParentChain()
(lang:getNonEtymologicalCode() == self:getCode() and not self:hasAncestor(lang)) -- Etymology-only child which isn't an ancestor.
for _, ancestorParent in ipairs(ancestorParents) do
)
if ancestorParent:getCode() == self._code or ancestorParent:hasAncestor(ancestor) then
) then
break
if format == "object" then
else
insert(descendants, lang)
insert(ancestorsParents, ancestorParent)
elseif format == "code" then
insert(descendants, code)
elseif format == "name" then
insert(descendants, lang:getCanonicalName())
end
end
end
end
end
end
end
for _, ancestorParent in ipairs(ancestorsParents) do
return descendants
local ret = func(ancestorParent)
end
if ret then return ret end
function Language:getDescendants()
if not self._descendantObjects then
self._descendantObjects = fetch_descendants(self, "object")
end
end
return self._descendantObjects
end
end

local function do_iteration(otherlang, parent_check)
function Language:getDescendantCodes()
-- otherlang can't be self
if not self._descendantCodes then
if (type(otherlang) == "string" and otherlang or otherlang:getCode()) == self._code then
self._descendantCodes = fetch_descendants(self, "code")
return false
end
end
repeat
return self._descendantCodes
if iterateOverAncestorTree(
self,
function(ancestor)
return ancestor:getCode() == (type(otherlang) == "string" and otherlang or otherlang:getCode())
end,
parent_check
) then
return true
elseif type(otherlang) == "string" then
otherlang = export.getByCode(otherlang, nil, true, nil, self._useRequire)
end
otherlang = otherlang:getParent()
parent_check = false
until not otherlang
end
end

local parent_check = true
function Language:getDescendantNames()
for _, otherlang in ipairs{...} do
if not self._descendantNames then
local ret = do_iteration(otherlang, parent_check)
self._descendantNames = fetch_descendants(self, "name")
if ret then
return true
end
end
return self._descendantNames
end
end
return false
end
function Language:hasDescendant(...)

for _, lang in ipairs{...} do
function Language:getAncestorChain()
if type(lang) == "string" then
if not self._ancestorChain then
lang = export.getByCode(lang, nil, true, nil, useRequire)
self._ancestorChain = {}
end
local step = self
if lang:hasAncestor(self) then
return true
while true do
local ancestors = step:getAncestors()
end
step = #ancestors == 1 and ancestors[1] or nil
if not step then break end
insert(self._ancestorChain, 1, step)
end
end
return false
end
end
return self._ancestorChain
end
local function fetch_children(self, format)

local m_etym_data = require("Module:etymology languages/data")
local function fetch_descendants(self, format)
local self_code = self:getCode()
local languages = require("Module:languages/code to canonical name")
local children = {}
local etymology_languages = require("Module:etymology languages/code to canonical name")
for code, data in pairs(m_etym_data) do
local families = require("Module:families/code to canonical name")
local _data = data
local descendants = {}
repeat
local family = self:getFamily()
parent = _data[5]
-- Iterate over all three datasets.
if parent == self_code then
for _, data in ipairs{languages, etymology_languages, families} do
if format == "object" then
for code in pairs(data) do
insert(children, export.getByCode(code, nil, true, nil, useRequire))
local lang = export.getByCode(code, nil, true, true, self._useRequire)
elseif format == "code" then
-- Test for a descendant. Earlier tests weed out most candidates, while the more intensive tests are only used sparingly.
insert(children, code)
if (
elseif format == "name" then
code ~= self._code and -- Not self.
insert(children, data[1])
lang:inFamily(family) and -- In the same family.
end
break
(
family:getProtoLanguageCode() == self._code or -- Self is the protolanguage.
self:hasDescendant(lang) or -- Full hasDescendant check.
(lang:getFullCode() == self._code and not self:hasAncestor(lang)) -- Etymology-only child which isn't an ancestor.
)
) then
if format == "object" then
insert(descendants, lang)
elseif format == "code" then
insert(descendants, code)
elseif format == "name" then
insert(descendants, lang:getCanonicalName())
end
end
end
_data = m_etym_data[parent]
until not _data
end
end
return children
end
end
return descendants
end
function Language:getChildren()

if not self._childObjects then
function Language:getDescendants()
self._childObjects = fetch_children(self, "object")
if not self._descendantObjects then
end
self._descendantObjects = fetch_descendants(self, "object")
return self._childObjects
end
end
return self._descendantObjects
end
function Language:getChildrenCodes()

if not self._childCodes then
function Language:getDescendantCodes()
self._childCodes = fetch_children(self, "code")
if not self._descendantCodes then
end
self._descendantCodes = fetch_descendants(self, "code")
return self._childCodes
end
end
return self._descendantCodes
end
function Language:getChildrenNames()

if not self._childNames then
function Language:getDescendantNames()
self._childNames = fetch_children(self, "name")
if not self._descendantNames then
end
self._descendantNames = fetch_descendants(self, "name")
return self._childNames
end
end
return self._descendantNames
end
function Language:hasChild(...)

local lang = ...
function Language:hasDescendant(...)
if not lang then
for _, lang in ipairs{...} do
return false
elseif type(lang) == "string" then
if type(lang) == "string" then
lang = export.getByCode(lang, nil, true, nil, useRequire)
lang = export.getByCode(lang, nil, true, nil, self._useRequire)
end
end
if lang:hasParent(self) then
if lang:hasAncestor(self) then
return true
return true
end
end
return self:hasChild(select(2, ...))
end
end
return false
end
--[==[Returns the name of the main category of that language. Example: {{code|lua|"French language"}} for French, whose category is at [[:Category:French language]]. Unless optional argument <code>nocap</code> is given, the language name at the beginning of the returned value will be capitalized. This capitalization is correct for category names, but not if the language name is lowercase and the returned value of this function is used in the middle of a sentence.]==]

function Language:getCategoryName(nocap)
local function fetch_children(self, format)
if not self._categoryName then
local m_etym_data = require("Module:etymology languages/data")
local name = self:getCanonicalName()
local self_code = self._code
-- Only add " language" if a regular language.
local children = {}
if #self._stack == 1 then
for code, data in pairs(m_etym_data) do
-- If the name already has "language" in it, don't add it.
local _data = data
if not name:match("[Ll]anguage$") then
repeat
name = name .. " language"
local parent = _data[5]
if parent == self_code then
if format == "object" then
insert(children, export.getByCode(code, nil, true, nil, self._useRequire))
elseif format == "code" then
insert(children, code)
elseif format == "name" then
insert(children, data[1])
end
end
break
end
end
_data = m_etym_data[parent]
self._categoryName = name
until not _data
end
if nocap then
return self._categoryName
else
return mw.getContentLanguage():ucfirst(self._categoryName)
end
end
end
return children
end
--[==[Creates a link to the category; the link text is the canonical name.]==]

function Language:makeCategoryLink()
function Language:getChildren()
return "[[:Category:" .. self:getCategoryName() .. "|" .. self:getDisplayForm() .. "]]"
if not self._childObjects then
self._childObjects = fetch_children(self, "object")
end
end
return self._childObjects
end
function Language:getStandardCharacters(sc)

if type(self._rawData.standardChars) ~= "table" then
function Language:getChildrenCodes()
return self._rawData.standardChars
if not self._childCodes then
else
self._childCodes = fetch_children(self, "code")
if sc and type(sc) ~= "string" then
end
checkObject("script", nil, sc)
return self._childCodes
sc = sc:getCode()
end
end

if (not sc) or sc == "None" then
function Language:getChildrenNames()
local scripts = {}
if not self._childNames then
for _, script in pairs(self._rawData.standardChars) do
self._childNames = fetch_children(self, "name")
insert(scripts, script)
end
end
return table.concat(scripts)
return self._childNames
end
end

if self._rawData.standardChars[sc] then
function Language:hasChild(...)
return self._rawData.standardChars[sc] .. (self._rawData.standardChars[1] or "")
local lang = ...
if not lang then
return false
elseif type(lang) == "string" then
lang = export.getByCode(lang, nil, true, nil, self._useRequire)
end
if lang:hasParent(self) then
return true
end
return self:hasChild(select(2, ...))
end

--[==[Returns the name of the main category of that language. Example: {{code|lua|"French language"}} for French, whose category is at [[:Category:French language]]. Unless optional argument <code>nocap</code> is given, the language name at the beginning of the returned value will be capitalized. This capitalization is correct for category names, but not if the language name is lowercase and the returned value of this function is used in the middle of a sentence.]==]
function Language:getCategoryName(nocap)
if not self._categoryName then
local name = self:getCanonicalName()
-- Only add " language" if a full language.
if #self._stack == 1 then
-- If the name already has "language" in it, don't add it.
if not name:match("[Ll]anguage$") then
name = name .. " language"
end
end
end
end
self._categoryName = name
end
end
if nocap then
return self._categoryName
--[==[Make the entry name (i.e. the correct page name).]==]
else
function Language:makeEntryName(text, sc)
return mw.getContentLanguage():ucfirst(self._categoryName)
if (not text) or text == "" then
end
return text, nil, {}
end

--[==[Creates a link to the category; the link text is the canonical name.]==]
function Language:makeCategoryLink()
return "[[:Category:" .. self:getCategoryName() .. "|" .. self:getDisplayForm() .. "]]"
end

function Language:getStandardCharacters(sc)
if type(self._rawData.standardChars) ~= "table" then
return self._rawData.standardChars
else
if sc and type(sc) ~= "string" then
check_object("script", nil, sc)
sc = sc:getCode()
end
end
if (not sc) or sc == "None" then
local scripts = {}
-- Remove directional characters.
for _, script in pairs(self._rawData.standardChars) do
text = remove_directional_chars(text)
insert(scripts, script)
-- Set `unsupported` as true if certain conditions are met.
local unsupported
-- Check if there's an unsupported character. \239\191\189 is the replacement character U+FFFD, which can't be typed directly here due to an abuse filter. Unix-style dot-slash notation is also unsupported, as it is used for relative paths in links, as are 3 or more consecutive tildes.
if (
text:find("[#<>%[%]_{|}]") or
text:find("\239\191\189") or
text:find("%f[^%z/]%.%.?%f[%z/]") or
text:find("~~~")
) then
unsupported = true
-- If it looks like an interwiki link.
elseif text:find(":") then
local prefix = text:gsub("^:*(.-):.*", string.ulower)
if (
conditionalRequire("Module:data/namespaces")[prefix] or
conditionalRequire("Module:data/interwikis")[prefix]
) then
unsupported = true
end
end
return concat(scripts)
end
end
if self._rawData.standardChars[sc] then
return self._rawData.standardChars[sc] .. (self._rawData.standardChars[1] or "")
-- Check if the text is a listed unsupported title.
local unsupportedTitles = conditionalRequire("Module:links/data").unsupported_titles
if unsupportedTitles[text] then
return "Unsupported titles/" .. unsupportedTitles[text], nil, {}
end
sc = checkScript(text, self, sc)
local fail, cats
text = normalize(text, sc)
text, fail, cats = iterateSectionSubstitutions(text, nil, nil, self, sc, self._rawData.entry_name, "makeEntryName")
text = ugsub(text, "^[¿¡]?([^%s%p]+)%s*[؟?!;՛՜ ՞ ՟？！︖︕।॥။၊་།]?$", "%1") or text
-- Escape unsupported characters so they can be used in titles. ` is used as a delimiter for this, so a raw use of it in an unsupported title is also escaped here to prevent interference; this is only done with unsupported titles, though, so inclusion won't in itself mean a title is treated as unsupported (which is why it's excluded from the earlier test).
if unsupported then
local unsupported_characters = conditionalRequire("Module:links/data").unsupported_characters
text = text
:gsub("[#<>%[%]_`{|}\239]\191?\189?", unsupported_characters)
:gsub("%f[^%z/]%.%.?%f[%z/]", function(m)
return m:gsub("%.", "`period`")
end)
:gsub("~~~+", function(m)
return m:gsub("~", "`tilde`")
end)
text = "Unsupported titles/" .. text
end
end
return text, fail, cats
end
end
end

--[==[Generates alternative forms using a specified method, and returns them as a table. If no method is specified, returns a table containing only the input term.]==]
--[==[Make the entry name (i.e. the correct page name).]==]
function Language:generateForms(text, sc)
function Language:makeEntryName(text, sc)
if self._rawData.generate_forms then
if (not text) or text == "" then
sc = checkScript(text, self, sc)
return text, nil, {}
return require("Module:" .. self._rawData.generate_forms).generateForms(text, self:getCode(), sc:getCode())
else
return {text}
end
end
end
-- Set `unsupported` as true if certain conditions are met.
--[==[Creates a sort key for the given entry name, following the rules appropriate for the language. This removes diacritical marks from the entry name if they are not considered significant for sorting, and may perform some other changes. Any initial hyphen is also removed, and anything parentheses is removed as well.
local unsupported
The <code>sort_key</code> setting for each language in the data modules defines the replacements made by this function, or it gives the name of the module that takes the entry name and returns a sortkey.]==]
-- Check if there's an unsupported character. \239\191\189 is the replacement character U+FFFD, which can't be typed directly here due to an abuse filter. Unix-style dot-slash notation is also unsupported, as it is used for relative paths in links, as are 3 or more consecutive tildes.
function Language:makeSortKey(text, sc)
if (
if (not text) or text == "" then
return text, nil, {}
text:find("[#<>%[%]_{|}]") or
text:find("\239\191\189") or
end
if text:find("<[^<>]+>") then
text:find("%f[^%z/]%.%.?%f[%z/]") or
track("track HTML tag")
text:find("~~~")
) then
end
unsupported = true
-- Remove directional characters, soft hyphens, strip markers and HTML tags.
-- If it looks like an interwiki link.
text = ugsub(text, "[\194\173" .. dir_char .. "]", "")
elseif text:find(":") then
text = mw.text.unstrip(text)
:gsub("<[^<>]+>", "")
local prefix = text:gsub("^:*(.-):.*", string.ulower)
if (
self:loadData("Module:data/namespaces")[prefix] or
text = mw.uri.decode(text, "PATH")
self:loadData("Module:data/interwikis")[prefix]
text = checkNoEntities(text)
) then
unsupported = true
-- Remove initial hyphens and * unless the term only consists of spacing + punctuation characters.
text = ugsub(text, "^([􀀀-􏿽]*)[-־ـ᠊*]+([􀀀-􏿽]*)(.*[^%s%p].*)", "%1%2%3")
sc = checkScript(text, self, sc)
text = normalize(text, sc)
text = removeCarets(text, sc)
-- For languages with dotted dotless i, ensure that "İ" is sorted as "i", and "I" is sorted as "ı".
if self:hasDottedDotlessI() then
text = text
:gsub(mw.ustring.toNFD("İ"), "i")
:gsub("I", "ı")
text = sc:toFixedNFD(text)
end
-- Convert to lowercase, make the sortkey, then convert to uppercase. Where the language has dotted dotless i, it is usually not necessary to convert "i" to "İ" and "ı" to "I" first, because "I" will always be interpreted as conventional "I" (not dotless "İ") by any sorting algorithms, which will have been taken into account by the sortkey substitutions themselves. However, if no sortkey substitutions have been specified, then conversion is necessary so as to prevent "i" and "ı" both being sorted as "I".
-- An exception is made for scripts that (sometimes) sort by scraping page content, as that means they are sensitive to changes in capitalization (as it changes the target page).
local fail, cats
if not sc:sortByScraping() then
text = text:ulower()
end
end
end

text, fail, cats = iterateSectionSubstitutions(text, nil, nil, self, sc, self._rawData.sort_key, "makeSortKey")
-- Check if the text is a listed unsupported title.
local unsupportedTitles = self:loadData("Module:links/data").unsupported_titles
if not sc:sortByScraping() then
if unsupportedTitles[text] then
if self:hasDottedDotlessI() and not self._rawData.sort_key then
return "Unsupported titles/" .. unsupportedTitles[text], nil, {}
text = text
end
:gsub("ı", "I")

:gsub("i", "İ")
text = sc:toFixedNFC(text)
sc = checkScript(text, self, sc)

end
local fail, cats
text = text:uupper()
text = normalize(text, sc)
end
text, fail, cats = iterateSectionSubstitutions(text, nil, nil, self, sc, self._rawData.entry_name, "makeEntryName")

-- Remove parentheses, as long as they are either preceded or followed by something.
text = umatch(text, "^[¿¡]?(.-[^%s%p].-)%s*[؟?!;՛՜ ՞ ՟？！︖︕।॥။၊་།]?$") or text


-- Escape unsupported characters so they can be used in titles. ` is used as a delimiter for this, so a raw use of it in an unsupported title is also escaped here to prevent interference; this is only done with unsupported titles, though, so inclusion won't in itself mean a title is treated as unsupported (which is why it's excluded from the earlier test).
if unsupported then
local unsupported_characters = self:loadData("Module:links/data").unsupported_characters
text = text
text = text
:gsub("(.)[()]+", "%1")
:gsub("[#<>%[%]_`{|}\239]\191?\189?", unsupported_characters)
:gsub("[()]+(.)", "%1")
:gsub("%f[^%z/]%.%.?%f[%z/]", function(m)
return m:gsub("%.", "`period`")
end)
text = escape_risky_characters(text)
:gsub("~~~+", function(m)
return text, fail, cats
return m:gsub("~", "`tilde`")
end)
text = "Unsupported titles/" .. text
end
end

return text, fail, cats
--[==[Create the form used as as a basis for display text and transliteration.]==]
end
local function processDisplayText(text, self, sc, keepCarets, keepPrefixes)

local subbedChars = {}
--[==[Generates alternative forms using a specified method, and returns them as a table. If no method is specified, returns a table containing only the input term.]==]
text, subbedChars = doTempSubstitutions(text, subbedChars, keepCarets)
function Language:generateForms(text, sc)
if self._rawData.generate_forms then
text = mw.uri.decode(text, "PATH")
text = checkNoEntities(text)
sc = checkScript(text, self, sc)
sc = checkScript(text, self, sc)
return require("Module:" .. self._rawData.generate_forms).generateForms(text, self._code, sc:getCode())
local fail, cats
else
text = normalize(text, sc)
return {text}
text, fail, cats, subbedChars = iterateSectionSubstitutions(text, subbedChars, keepCarets, self, sc, self._rawData.display_text, "makeDisplayText")
text = removeCarets(text, sc)
-- Remove any interwiki link prefixes (unless they have been escaped or this has been disabled).
if text:match(":") and not keepPrefixes then
local rep
repeat
text, rep = text:gsub("\\\\(\\*:)", "\3%1")
until rep == 0
text = text
:gsub("\\:", "\4")
while true do
local prefix = text:gsub("^(.-):.+", function(m1)
return m1:gsub("\244[\128-\191]*", "")
end)
if not prefix or prefix == text then
break
end
local lower_prefix = prefix:ulower()
if not (conditionalRequire("Module:data/interwikis")[lower_prefix] or prefix == "") then
break
end
text = text:gsub("^(.-):(.*)", function(m1, m2)
local ret = {}
for subbedChar in m1:gmatch("\244[\128-\191]*") do
insert(ret, subbedChar)
end
return table.concat(ret) .. m2
end)
end
text = text
:gsub("\3", "\\")
:gsub("\4", ":")
end
return text, fail, cats, subbedChars
end
end
end

--[==[Make the display text (i.e. what is displayed on the page).]==]
--[==[Creates a sort key for the given entry name, following the rules appropriate for the language. This removes diacritical marks from the entry name if they are not considered significant for sorting, and may perform some other changes. Any initial hyphen is also removed, and anything parentheses is removed as well.
function Language:makeDisplayText(text, sc, keepPrefixes)
The <code>sort_key</code> setting for each language in the data modules defines the replacements made by this function, or it gives the name of the module that takes the entry name and returns a sortkey.]==]
if (not text) or text == "" then
function Language:makeSortKey(text, sc)
return text, nil, {}
if (not text) or text == "" then
end
return text, nil, {}
text = remove_directional_chars(text)
local fail, cats, subbedChars
text, fail, cats, subbedChars = processDisplayText(text, self, sc, nil, keepPrefixes)
text = escape_risky_characters(text)
return undoTempSubstitutions(text, subbedChars), fail, cats
end
end
if text:find("<[^<>]+>") then
track("track HTML tag")
--[==[Transliterates the text from the given script into the Latin script (see [[Wiktionary:Transliteration and romanization]]). The language must have the <code>translit</code> property for this to work; if it is not present, {{code|lua|nil}} is returned.
end
Returns three values:
-- Remove directional characters, soft hyphens, strip markers and HTML tags.
# The transliteration.
text = ugsub(text, "[\194\173\226\128\170-\226\128\174\226\129\166-\226\129\169]", "")
# A boolean which indicates whether the transliteration failed for an unexpected reason. If {{code|lua|false}}, then the transliteration either succeeded, or the module is returning nothing in a controlled way (e.g. the input was {{code|lua|"-"}}). Generally, this means that no maintenance action is required. If {{code|lua|true}}, then the transliteration is {{code|lua|nil}} because either the input or output was defective in some way (e.g. [[Module:ar-translit]] will not transliterate non-vocalised inputs, and this module will fail partially-completed transliterations in all languages). Note that this value can be manually set by the transliteration module, so make sure to cross-check to ensure it is accurate.
text = mw.text.unstrip(text)
# A table of categories selected by the transliteration module, which should be in the format expected by {{code|lua|format_categories}} in [[Module:utilities]].
:gsub("<[^<>]+>", "")
The <code>sc</code> parameter is handled by the transliteration module, and how it is handled is specific to that module. Some transliteration modules may tolerate {{code|lua|nil}} as the script, others require it to be one of the possible scripts that the module can transliterate, and will show an error if it's not one of them. For this reason, the <code>sc</code> parameter should always be provided when writing non-language-specific code.

The <code>module_override</code> parameter is used to override the default module that is used to provide the transliteration. This is useful in cases where you need to demonstrate a particular module in use, but there is no default module yet, or you want to demonstrate an alternative version of a transliteration module before making it official. It should not be used in real modules or templates, only for testing. All uses of this parameter are tracked by [[Template:tracking/module_override]].
text = decode_uri(text, "PATH")
'''Known bugs''':
text = checkNoEntities(self, text)
* This function assumes {tr(s1) .. tr(s2) == tr(s1 .. s2)}. When this assertion fails, wikitext markups like <nowiki>'''</nowiki> can cause wrong transliterations.

* HTML entities like <code>&amp;apos;</code>, often used to escape wikitext markups, do not work.]==]
-- Remove initial hyphens and * unless the term only consists of spacing + punctuation characters.
function Language:transliterate(text, sc, module_override)
text = ugsub(text, "^([􀀀-􏿽]*)[-־ـ᠊*]+([􀀀-􏿽]*)(.*[^%s%p].*)", "%1%2%3")
-- If there is no text, or the language doesn't have transliteration data and there's no override, return nil.

if not (self._rawData.translit or module_override) then
sc = checkScript(text, self, sc)
return nil, false, {}

elseif (not text) or text == "" or text == "-" then
text = normalize(text, sc)
return text, false, {}
text = removeCarets(text, sc)

-- For languages with dotted dotless i, ensure that "İ" is sorted as "i", and "I" is sorted as "ı".
if self:hasDottedDotlessI() then
text = text:gsub("I\204\135", "i") -- decomposed "İ"
:gsub("I", "ı")
text = sc:toFixedNFD(text)
end
-- Convert to lowercase, make the sortkey, then convert to uppercase. Where the language has dotted dotless i, it is usually not necessary to convert "i" to "İ" and "ı" to "I" first, because "I" will always be interpreted as conventional "I" (not dotless "İ") by any sorting algorithms, which will have been taken into account by the sortkey substitutions themselves. However, if no sortkey substitutions have been specified, then conversion is necessary so as to prevent "i" and "ı" both being sorted as "I".
-- An exception is made for scripts that (sometimes) sort by scraping page content, as that means they are sensitive to changes in capitalization (as it changes the target page).
local fail, cats
if not sc:sortByScraping() then
text = text:ulower()
end

text, fail, cats = iterateSectionSubstitutions(text, nil, nil, self, sc, self._rawData.sort_key, "makeSortKey")

if not sc:sortByScraping() then
if self:hasDottedDotlessI() and not self._rawData.sort_key then
text = text:gsub("ı", "I")
:gsub("i", "İ")
text = sc:toFixedNFC(text)
end
end
text = text:uupper()
-- If the script is not transliteratable (and no override is given), return nil.
end
sc = checkScript(text, self, sc)

if not (sc:isTransliterated() or module_override) then
-- Remove parentheses, as long as they are either preceded or followed by something.
return nil, true, {}
text = text
end
:gsub("(.)[()]+", "%1")
:gsub("[()]+(.)", "%1")
-- Remove any strip markers and directional characters.

text = mw.text.unstrip(text)
text = remove_directional_chars(text)
text = escape_risky_characters(text)
return text, fail, cats
end
-- Get the display text with the keepCarets flag set.

local fail, cats, subbedChars
--[==[Create the form used as as a basis for display text and transliteration.]==]
text, fail, cats, subbedChars = processDisplayText(text, self, sc, true)
local function processDisplayText(text, self, sc, keepCarets, keepPrefixes)
local subbedChars = {}
-- Transliterate (using the module override if applicable).
text, fail, cats, subbedChars = iterateSectionSubstitutions(text, subbedChars, true, self, sc, module_override or self._rawData.translit, "tr")
text, subbedChars = doTempSubstitutions(text, subbedChars, keepCarets)

text = decode_uri(text, "PATH")
-- Incomplete transliterations return nil.
text = checkNoEntities(self, text)
if text then

if sc:countCharacters(text) > 0 then
sc = checkScript(text, self, sc)
-- Remove any characters in Latin, which includes Latin characters also included in other scripts (as these are false positives). Anything remaining should only be script code "None" (e.g. numerals).
local fail, cats
local check_text = ugsub(text, "[" .. require("Module:scripts").getByCode("Latn"):getCharacters() .. "]", "")
text = normalize(text, sc)
if require("Module:scripts").findBestScriptWithoutLang(check_text) ~= "None" then
text, fail, cats, subbedChars = iterateSectionSubstitutions(text, subbedChars, keepCarets, self, sc, self._rawData.display_text, "makeDisplayText")
return nil, true, cats

end
text = removeCarets(text, sc)

-- Remove any interwiki link prefixes (unless they have been escaped or this has been disabled).
if text:match(":") and not keepPrefixes then
local rep
repeat
text, rep = text:gsub("\\\\(\\*:)", "\3%1")
until rep == 0
text = text
:gsub("\\:", "\4")
while true do
local prefix = text:gsub("^(.-):.+", function(m1)
return m1:gsub("\244[\128-\191]*", "")
end)
if not prefix or prefix == text then
break
end
end
local lower_prefix = prefix:ulower()
else
if not (self:loadData("Module:data/interwikis")[lower_prefix] or prefix == "") then
return nil, true, cats
end
break
end
text = text:gsub("^(.-):(.*)", function(m1, m2)
text = escape_risky_characters(text)
local ret = {}
text = undoTempSubstitutions(text, subbedChars)
for subbedChar in gmatch(m1, "\244[\128-\191]*") do
insert(ret, subbedChar)
-- If the script does not use capitalization, then capitalize any letters of the transliteration which are immediately preceded by a caret (and remove the caret).
end
if text and not sc:hasCapitalization() and text:match("%^") then
return concat(ret) .. m2
text = processCarets(text, "%^([\128-\191\244]*%*?)([^\128-\191\244][\128-\191]*)", function(m1, m2)
return m1 .. m2:uupper()
end)
end)
end
end
text = text
:gsub("\3", "\\")
-- Track module overrides.
:gsub("\4", ":")
if module_override ~= nil then
track("module_override")
end
fail = text == nil and (not not fail) or false
return text, fail, cats
end
end

return text, fail, cats, subbedChars
function Language:overrideManualTranslit()
end
return not not self._rawData.override_translit

--[==[Make the display text (i.e. what is displayed on the page).]==]
function Language:makeDisplayText(text, sc, keepPrefixes)
if (not text) or text == "" then
return text, nil, {}
end
end
local fail, cats, subbedChars
--[==[Returns {{code|lua|true}} if the language has a transliteration module, or {{code|lua|false}} if it doesn't.]==]
text, fail, cats, subbedChars = processDisplayText(text, self, sc, nil, keepPrefixes)
function Language:hasTranslit()

return not not self._rawData.translit
text = escape_risky_characters(text)
return undoTempSubstitutions(text, subbedChars), fail, cats
end

--[==[Transliterates the text from the given script into the Latin script (see [[Wiktionary:Transliteration and romanization]]). The language must have the <code>translit</code> property for this to work; if it is not present, {{code|lua|nil}} is returned.
Returns three values:
# The transliteration.
# A boolean which indicates whether the transliteration failed for an unexpected reason. If {{code|lua|false}}, then the transliteration either succeeded, or the module is returning nothing in a controlled way (e.g. the input was {{code|lua|"-"}}). Generally, this means that no maintenance action is required. If {{code|lua|true}}, then the transliteration is {{code|lua|nil}} because either the input or output was defective in some way (e.g. [[Module:ar-translit]] will not transliterate non-vocalised inputs, and this module will fail partially-completed transliterations in all languages). Note that this value can be manually set by the transliteration module, so make sure to cross-check to ensure it is accurate.
# A table of categories selected by the transliteration module, which should be in the format expected by {{code|lua|format_categories}} in [[Module:utilities]].
The <code>sc</code> parameter is handled by the transliteration module, and how it is handled is specific to that module. Some transliteration modules may tolerate {{code|lua|nil}} as the script, others require it to be one of the possible scripts that the module can transliterate, and will show an error if it's not one of them. For this reason, the <code>sc</code> parameter should always be provided when writing non-language-specific code.
The <code>module_override</code> parameter is used to override the default module that is used to provide the transliteration. This is useful in cases where you need to demonstrate a particular module in use, but there is no default module yet, or you want to demonstrate an alternative version of a transliteration module before making it official. It should not be used in real modules or templates, only for testing. All uses of this parameter are tracked by [[Wiktionary:Tracking/module_override]].
'''Known bugs''':
* This function assumes {tr(s1) .. tr(s2) == tr(s1 .. s2)}. When this assertion fails, wikitext markups like <nowiki>'''</nowiki> can cause wrong transliterations.
* HTML entities like <code>&amp;apos;</code>, often used to escape wikitext markups, do not work.]==]
function Language:transliterate(text, sc, module_override)
-- If there is no text, or the language doesn't have transliteration data and there's no override, return nil.
if not (self._rawData.translit or module_override) then
return nil, false, {}
elseif (not text) or text == "" or text == "-" then
return text, false, {}
end
end
-- If the script is not transliteratable (and no override is given), return nil.
sc = checkScript(text, self, sc)
function Language:link_tr()
if not (sc:isTransliterated() or module_override) then
return not not self._rawData.link_tr
return nil, true, {}
end
end

-- Remove any strip markers.
text = mw.text.unstrip(text)

-- Get the display text with the keepCarets flag set.
local fail, cats, subbedChars
text, fail, cats, subbedChars = processDisplayText(text, self, sc, true)

-- Transliterate (using the module override if applicable).
text, fail, cats, subbedChars = iterateSectionSubstitutions(text, subbedChars, true, self, sc, module_override or self._rawData.translit, "tr")
if not text then
--[==[Returns {{code|lua|true}} if the language uses the letters I/ı and İ/i, or {{code|lua|false}} if it doesn't.]==]
return nil, true, cats
function Language:hasDottedDotlessI()
return not not self._rawData.dotted_dotless_i
end
end
-- Incomplete transliterations return nil.
function Language:toJSON(returnTable)
local entryNamePatterns = nil
local charset = sc.characters
if charset and umatch(text, "[" .. charset .. "]") then
local entryNameRemoveDiacritics = nil
-- Remove any characters in Latin, which includes Latin characters also included in other scripts (as these are false positives). Anything remaining should only be script code "None" (e.g. numerals).
local check_text = ugsub(text, "[" .. require("Module:scripts").getByCode("Latn").characters .. "]", "")
if self._rawData.entry_name then
if require("Module:scripts").findBestScriptWithoutLang(check_text):getCode() ~= "None" then
entryNameRemoveDiacritics = self._rawData.entry_name.remove_diacritics
return nil, true, cats
if self._rawData.entry_name.from then
entryNamePatterns = {}
for i, from in ipairs(self._rawData.entry_name.from) do
insert(entryNamePatterns, {from = from, to = self._rawData.entry_name.to[i] or ""})
end
end
end
if not self._type then
self:hasType()
end
local types = {}
for type in pairs(self._type) do
insert(types, type)
end
end
end


text = escape_risky_characters(text)
-- mainCode should only end up non-nil if dontCanonicalizeAliases is passed to make_object().
text = undoTempSubstitutions(text, subbedChars)
local ret = {

ancestors = self:getAncestorCodes(),
-- If the script does not use capitalization, then capitalize any letters of the transliteration which are immediately preceded by a caret (and remove the caret).
canonicalName = self:getCanonicalName(),
if text and not sc:hasCapitalization() and text:match("%^") then
categoryName = self:getCategoryName("nocap"),
text = processCarets(text, "%^([\128-\191\244]*%*?)([^\128-\191\244][\128-\191]*)", function(m1, m2)
code = self:getCode(),
return m1 .. m2:uupper()
mainCode = self._main_code,
end)
entryNamePatterns = entryNamePatterns,
entryNameRemoveDiacritics = entryNameRemoveDiacritics,
family = self:getFamilyCode(),
otherNames = self:getOtherNames(true),
aliases = self:getAliases(),
varieties = self:getVarieties(),
scripts = self:getScriptCodes(),
parent = self._parentCode or nil,
nonEtymological = self._nonEtymologicalCode or nil,
type = types,
wikimediaLanguages = self:getWikimediaLanguageCodes(),
wikidataItem = self:getWikidataItem(),
}
ret = require("Module:table").deepcopy(ret)
if returnTable then
return ret
else
return require("Module:JSON").toJSON(ret)
end
end
end

-- Track module overrides.
--[==[
if module_override ~= nil then
<span style="color: #BA0000">This function is not for use in entries or other content pages.</span>
track("module_override")
Returns a blob of data about the language. The format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.
end
-- Do NOT use these methods!

-- All uses should be pre-approved on the talk page!
fail = text == nil and (not not fail) or false
]==]

function Language:getRawData()
return text, fail, cats
local rawData = {}
end
for _, element in ipairs(self._stack) do

for k, v in pairs(element) do
function Language:overrideManualTranslit()
rawData[k] = v
return not not self._rawData.override_translit
end

--[==[Returns {{code|lua|true}} if the language has a transliteration module, or {{code|lua|false}} if it doesn't.]==]
function Language:hasTranslit()
return not not self._rawData.translit
end

function Language:link_tr()
return not not self._rawData.link_tr
end

--[==[Returns {{code|lua|true}} if the language uses the letters I/ı and İ/i, or {{code|lua|false}} if it doesn't.]==]
function Language:hasDottedDotlessI()
return not not self._rawData.dotted_dotless_i
end

function Language:toJSON(returnTable)
local entryNamePatterns = nil
local entryNameRemoveDiacritics = nil

if self._rawData.entry_name then
entryNameRemoveDiacritics = self._rawData.entry_name.remove_diacritics
if self._rawData.entry_name.from then
entryNamePatterns = {}
for i, from in ipairs(self._rawData.entry_name.from) do
insert(entryNamePatterns, {from = from, to = self._rawData.entry_name.to[i] or ""})
end
end
end
end
return rawData
end
end
-- mainCode should only end up non-nil if dontCanonicalizeAliases is passed to make_object().
--[==[<span style="color: #BA0000">This function is not for use in entries or other content pages.</span>
local ret = m_table.deepcopy{
Returns a blob of data about the language that contains the "extra data". Much like with getRawData, the format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.]==]
ancestors = self:getAncestorCodes(),
function Language:getRawExtraData()
canonicalName = self:getCanonicalName(),
if #self._stack == 1 then
categoryName = self:getCategoryName("nocap"),
self:loadInExtraData()
code = self._code,
mainCode = self._main_code,
entryNamePatterns = entryNamePatterns,
entryNameRemoveDiacritics = entryNameRemoveDiacritics,
family = self:getFamilyCode(),
otherNames = self:getOtherNames(true),
aliases = self:getAliases(),
varieties = self:getVarieties(),
scripts = self:getScriptCodes(),
parent = self._parentCode or nil,
full = self._fullCode or nil,
type = m_table.keysToList(self:getTypes()),
wikimediaLanguages = self:getWikimediaLanguageCodes(),
wikidataItem = self:getWikidataItem(),
}

if returnTable then
return ret
else
return require("Module:JSON").toJSON(ret)
end
end

--[==[
<span style="color: #BA0000">This function is not for use in entries or other content pages.</span>
Returns a blob of data about the language. The format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.
-- Do NOT use these methods!
-- All uses should be pre-approved on the talk page!
]==]
function Language:getRawData()
local rawData = {}
for _, element in ipairs(self._stack) do
for k, v in pairs(element) do
rawData[k] = v
end
end
return self._extraData
end
end
return rawData
end
local function getRawExtraLanguageData(code)

local modulename = export.getExtraDataModuleName(code)
--[==[<span style="color: #BA0000">This function is not for use in entries or other content pages.</span>
return modulename and conditionalRequire("Module:" .. modulename)[code] or nil
Returns a blob of data about the language that contains the "extra data". Much like with getRawData, the format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.]==]
function Language:getRawExtraData()
if #self._stack == 1 then
self:loadInExtraData()
end
end
return self._extraData
end


function Language:loadInExtraData()
local function getRawExtraLanguageData(self, code)
local modulename = export.getExtraDataModuleName(code)
if not self._extraData then
return modulename and self:loadData("Module:" .. modulename)[code] or nil
-- load extra data from module and assign to _extraData field
end
-- use empty table as a fallback if extra data is nil

self._extraData = getRawExtraLanguageData(self:getCode()) or {}
function Language:loadInExtraData()
end
if not self._extraData then
-- load extra data from module and assign to _extraData field
-- use empty table as a fallback if extra data is nil
self._extraData = getRawExtraLanguageData(self, self._code) or {}
end
end
return Language
end
end


Line 1,403: Line 1,573:


do
do
local key_types = {
-- Data that isn't inherited from the parent.
[2] = "unique",
local function no_inherit(lang, t, k)
aliases = "unique",
if (
otherNames = "unique",
k == 2 or -- wikidata item
k == "aliases" or
type = "append",
k == "varieties" or
varieties = "unique"
}
k == "otherNames"

) then
return lang._stack[#lang._stack][k], true
end
end
-- Data that is appended by each generation.
local function append_data(lang, t, k)
if k == "type" then
local parts = {}
for i = 1, #lang._stack do
insert(parts, lang._stack[i][k])
end
if type(parts[1]) == "string" then
return table.concat(parts, ", "), true
end
return nil, true
end
end
local function inherit_data(lang, t, k)
local i = #lang._stack
while not lang._stack[i][k] and i > 1 do
i = i - 1
end
return lang._stack[i][k]
end
local function make_stack(code, input_code, data, parent, useRequire)
local function make_stack(code, input_code, data, parent, useRequire)
parent.__index = parent
parent.__index = parent

local lang = {_code = input_code}
local lang = {
_code = input_code,
_useRequire = useRequire or nil
}
-- This can only happen if dontCanonicalizeAliases is passed to make_object().
-- This can only happen if dontCanonicalizeAliases is passed to make_object().
if code ~= input_code then
if code ~= input_code then
lang._main_code = code
lang._main_code = code
end
end

-- Full language.
-- Full language.
if not parent._stack then
if not parent._stack then
-- Create stack, accessed with rawData metamethod.
-- Create stack, accessed with rawData metamethod.
lang._stack = parent._rawData and {parent._rawData, data} or {data}
local stack = parent._rawData and {parent._rawData, data} or {data}
lang._stack = stack
lang._rawData = setmetatable({}, {
lang._rawData = setmetatable({}, {
__index = function(t, k)
__index = function(t, k)
local key_type = key_types[k]
-- Data that isn't inherited from the parent.
if key_type == "unique" then
return stack[#stack][k]
-- Data that is appended by each generation.
elseif key_type == "append" then
local parts = {}
for i = 1, #stack do
insert(parts, stack[i][k])
end
if type(parts[1]) == "string" then
return concat(parts, ","), true
end
-- Otherwise, iterate down the stack, looking for a match.
-- Otherwise, iterate down the stack, looking for a match.
else
local ret, stop_now = no_inherit(lang, t, k)
if not stop_now then
local i = #stack
while not stack[i][k] and i > 1 do
ret, stop_now = append_data(lang, t, k)
end
i = i - 1
if not stop_now then
ret = inherit_data(lang, t, k)
end
if (
k == 4 or
k == "ancestors" or
k == "wikimedia_codes"
) then
if type(ret) == "table" then
return ret
elseif type(ret) == "string" then
return mw.text.split(ret, "%s*,%s*")
end
end
else
return stack[i][k]
return ret
end
end
end,
end,
Line 1,479: Line 1,627:
end
end
})
})
-- Non-etymological code is the parent code.
-- Full code is the parent code.
lang._nonEtymologicalCode = parent._code or code
lang._fullCode = parent._code or code
-- Etymology-only.
-- Etymology-only.
else
else
Line 1,487: Line 1,635:
lang._stack = parent._stack
lang._stack = parent._stack
insert(lang._stack, data)
insert(lang._stack, data)
-- Copy non-etymological code.
-- Copy full code.
lang._nonEtymologicalCode = parent._nonEtymologicalCode
lang._fullCode = parent._fullCode
end
end

return setmetatable(lang, parent)
return setmetatable(lang, parent)
end
end

function export.makeObject(code, data, useRequire, dontCanonicalizeAliases)
function export.makeObject(code, data, useRequire, dontCanonicalizeAliases)
if not data then
if not data then
return nil
return nil
end
end

-- Convert any aliases.
-- Convert any aliases.
local input_code = code
local input_code = code
code = normalize_code(code)
code = normalize_code(code)
input_code = dontCanonicalizeAliases and input_code or code
input_code = dontCanonicalizeAliases and input_code or code

if data.type:find("family") and not data[5] then
if data.type:find("family") and not data[5] then
return require("Module:families").makeObject(code, data, useRequire)
return require("Module:families").makeObject(code, data, useRequire)
Line 1,511: Line 1,659:
parent = export.getByCode(data[5], nil, true, true, useRequire)
parent = export.getByCode(data[5], nil, true, true, useRequire)
else
else
parent = make_language(code, data, useRequire)
parent = Language
end
end
return make_stack(code, input_code, data, parent, useRequire)
return make_stack(code, input_code, data, parent, useRequire)
Line 1,518: Line 1,666:
end
end


--[==[Finds the language whose code matches the one provided. If it exists, it returns a <code class="nf">Language</code> object representing the language. Otherwise, it returns {{code|lua|nil}}, unless <code class="n">paramForError</code> is given, in which case an error is generated. If <code class="n">paramForError</code> is {{code|lua|true}}, a generic error message mentioning the bad code is generated; otherwise <code class="n">paramForError</code> should be a string or number specifying the parameter that the code came from, and this parameter will be mentioned in the error message along with the bad code. If <code class="n">allowEtymLang</code> is specified, etymology language codes are allowed and looked up along with normal language codes. If <code class="n">allowFamily</code> is specified, language family codes are allowed and looked up along with normal language codes.]==]
--[==[Finds the language whose code matches the one provided. If it exists, it returns a <code class="nf">Language</code> object representing the language. Otherwise, it returns {{code|lua|nil}}, unless <code class="n">paramForError</code> is given, in which case an error is generated. If <code class="n">paramForError</code> is {{code|lua|true}}, a generic error message mentioning the bad code is generated; otherwise <code class="n">paramForError</code> should be a string or number specifying the parameter that the code came from, and this parameter will be mentioned in the error message along with the bad code. If <code class="n">allowEtymLang</code> is specified, etymology-only language codes are allowed and looked up along with normal language codes. If <code class="n">allowFamily</code> is specified, language family codes are allowed and looked up along with normal language codes.]==]
function export.getByCode(code, paramForError, allowEtymLang, allowFamily, useRequire)
function export.getByCode(code, paramForError, allowEtymLang, allowFamily, useRequire)
if type(code) ~= "string" then
if type(code) ~= "string" then
Line 1,524: Line 1,672:
if not code then
if not code then
typ = "nil"
typ = "nil"
elseif checkObject("language", true, code) then
elseif check_object("language", true, code) then
typ = "a language object"
typ = "a language object"
elseif checkObject("family", true, code) then
elseif check_object("family", true, code) then
typ = "a family object"
typ = "a family object"
else
else
Line 1,533: Line 1,681:
error("The function getByCode expects a string as its first argument, but received " .. typ .. ".")
error("The function getByCode expects a string as its first argument, but received " .. typ .. ".")
end
end

local function conditionalRequire(modulename)
local function conditionalRequire(modulename)
if useRequire then
if useRequire then
return require(modulename)
return require(modulename)
else
else
return mw.loadData(modulename)
return load_data(modulename)
end
end
end
end
Line 1,549: Line 1,697:
-- Codes duplicated been full and etymology-only languages
-- Codes duplicated been full and etymology-only languages
["bsg"] = true,
["bsg"] = true,
["nan-hok"] = true, -- full code: nan-hbl
["rdb"] = true,
["rdb"] = true,
["tgf"] = true,
["tgf"] = true,
-- Aliases actively being deprecated
["prv"] = true, -- oc-pro
-- Codes that will be converted to families
-- Codes that will be converted to families
["nan"] = true,
["nan"] = true,
-- Codes being renamed
["cmn-wadegile"] = true,
["wuu-ngb"] = true,
["wuu-hzh"] = true,
["wuu-szh"] = true,
}
}


Line 1,562: Line 1,716:
return true
return true
end
end

local modulename = export.getDataModuleName(code)
local modulename = export.getDataModuleName(code)

local function get_data(code)
local function get_data(code)
return modulename and
return modulename and
Line 1,572: Line 1,726:
(allowEtymLang and allowFamily and require("Module:families/track-bad-etym-code")(code) and conditionalRequire("Module:families/data/etymology")[code])
(allowEtymLang and allowFamily and require("Module:families/track-bad-etym-code")(code) and conditionalRequire("Module:families/data/etymology")[code])
end
end

local data = get_data(code) or get_data(normalize_code(code))
local data = get_data(code) or get_data(normalize_code(code))

local retval = code and data and export.makeObject(code, data, useRequire)
local retval = code and data and export.makeObject(code, data, useRequire)

if not retval and paramForError then
if not retval and paramForError then
require("Module:languages/errorGetBy").code(code, paramForError, allowEtymLang, allowFamily)
require("Module:languages/errorGetBy").code(code, paramForError, allowEtymLang, allowFamily)
end
end

return retval
return retval
end
end


--[==[Finds the language whose canonical name (the name used to represent that language on Wiktionary) or other name matches the one provided. If it exists, it returns a <code class="nf">Language</code> object representing the language. Otherwise, it returns {{code|lua|nil}}, unless <code class="n">paramForError</code> is given, in which case an error is generated. If <code class="n">allowEtymLang</code> is specified, etymology language codes are allowed and looked up along with normal language codes. If <code class="n">allowFamily</code> is specified, language family codes are allowed and looked up along with normal language codes.
--[==[Finds the language whose canonical name (the name used to represent that language on Wiktionary) or other name matches the one provided. If it exists, it returns a <code class="nf">Language</code> object representing the language. Otherwise, it returns {{code|lua|nil}}, unless <code class="n">paramForError</code> is given, in which case an error is generated. If <code class="n">allowEtymLang</code> is specified, etymology-only language codes are allowed and looked up along with normal language codes. If <code class="n">allowFamily</code> is specified, language family codes are allowed and looked up along with normal language codes.
The canonical name of languages should always be unique (it is an error for two languages on Wiktionary to share the same canonical name), so this is guaranteed to give at most one result.
The canonical name of languages should always be unique (it is an error for two languages on Wiktionary to share the same canonical name), so this is guaranteed to give at most one result.
This function is powered by [[Module:languages/canonical names]], which contains a pre-generated mapping of non-etymology-language canonical names to codes. It is generated by going through the [[:Category:Language data modules]] for non-etymology languages. When <code class="n">allowEtymLang</code> is specified for the above function, [[Module:etymology languages/canonical names]] may also be used, and when <code class="n">allowFamily</code> is specified for the above function, [[Module:families/canonical names]] may also be used.]==]
This function is powered by [[Module:languages/canonical names]], which contains a pre-generated mapping of full-language canonical names to codes. It is generated by going through the [[:Category:Language data modules]] for full languages. When <code class="n">allowEtymLang</code> is specified for the above function, [[Module:etymology languages/canonical names]] may also be used, and when <code class="n">allowFamily</code> is specified for the above function, [[Module:families/canonical names]] may also be used.]==]
function export.getByCanonicalName(name, errorIfInvalid, allowEtymLang, allowFamily, useRequire)
function export.getByCanonicalName(name, errorIfInvalid, allowEtymLang, allowFamily, useRequire)
local function conditionalRequire(modulename)
local function conditionalRequire(modulename)
Line 1,592: Line 1,746:
return require(modulename)
return require(modulename)
else
else
return mw.loadData(modulename)
return load_data(modulename)
end
end
end
end

local byName = conditionalRequire("Module:languages/canonical names")
local byName = conditionalRequire("Module:languages/canonical names")
local code = byName and byName[name]
local code = byName and byName[name]

if not code and allowEtymLang then
if not code and allowEtymLang then
byName = conditionalRequire("Module:etymology languages/canonical names")
byName = conditionalRequire("Module:etymology languages/canonical names")
Line 1,609: Line 1,763:
byName[name:match("^(.*) languages$")]
byName[name:match("^(.*) languages$")]
end
end

if not code and allowFamily then
if not code and allowFamily then
byName = conditionalRequire("Module:families/canonical names")
byName = conditionalRequire("Module:families/canonical names")
Line 1,615: Line 1,769:
byName[name:match("^(.*) languages$")]
byName[name:match("^(.*) languages$")]
end
end

local retval = code and export.getByCode(code, errorIfInvalid, allowEtymLang, allowFamily, useRequire)
local retval = code and export.getByCode(code, errorIfInvalid, allowEtymLang, allowFamily, useRequire)

if not retval and errorIfInvalid then
if not retval and errorIfInvalid then
require("Module:languages/errorGetBy").canonicalName(name, allowEtymLang, allowFamily)
require("Module:languages/errorGetBy").canonicalName(name, allowEtymLang, allowFamily)
end
end

return retval
return retval
end
end
Line 1,627: Line 1,781:
--[==[Used by [[Module:languages/data/2]] (et al.) to add default types to the entities returned.]==]
--[==[Used by [[Module:languages/data/2]] (et al.) to add default types to the entities returned.]==]
function export.addDefaultTypes(data, regular, ...)
function export.addDefaultTypes(data, regular, ...)
local n = arg.n
for _, entity in pairs(data) do
local types = n > 0 and concat(arg, ",") or ""
for _, entity in next, data do
-- "regular" encompasses everything that doesn't have another type already assigned.
-- "regular" encompasses everything that doesn't have another type already assigned.
if regular then
if regular then
entity.type = entity.type or "regular"
entity.type = entity.type or "regular"
end
end
if n > 0 then
local types = table.concat({...}, ", ")
entity.type = types .. (entity.type and ("," .. entity.type) or "")
if #types > 0 then
end
entity.type = types .. (entity.type and (", " .. entity.type) or "")
end
return data
end

--[==[Used by [[Module:languages/data/2]] (et al.) and [[Module:etymology languages/data]] to finalize language-related data into the format that is actually returned.]==]
function export.finalizeLanguageData(data)
-- 4 is scripts.
local fields = {4, "ancestors", "type", "wikimedia_codes"}
local fields_len = #fields
for _, entity in next, data do
for i = 1, fields_len do
local key = fields[i]
local field = entity[key]
if field then
entity[key] = gsub(field, "%s+", "")
end
end
end
end
end
Line 1,640: Line 1,812:
end
end


--[==[Used by [[Module:etymology languages/data]] and [[Module:families/data/etymology]] to finalize the data into the format that is actually returned.]==]
--[==[Used by [[Module:etymology languages/data]] and [[Module:families/data/etymology]] to finalize etymology-related data into the format that is actually returned.]==]
function export.finalizeEtymologyData(data)
function export.finalizeEtymologyData(data)
local aliases = {}
local aliases = {}
for code, entity in pairs(data) do
for _, entity in next, data do
-- Move parent to 5 and family to 3.
-- Move parent to 5 and family to 3.
data[code][5] = data[code][3]
entity[5] = entity[3]
data[code][3] = data[code].family
entity[3] = entity.family
data[code].family = nil
entity.family = nil
end
end
for code, alias in pairs(aliases) do
for code, alias in next, aliases do
data[code] = alias
data[code] = alias
end
end
Line 1,655: Line 1,827:
end
end


--[==[For backwards compatibility only; modules should require the /error themselves.]==]
--[==[For backwards compatibility only; modules should require the error themselves.]==]
function export.err(lang_code, param, code_desc, template_tag, not_real_lang)
function export.err(lang_code, param, code_desc, template_tag, not_real_lang)
return require("Module:languages/error")(lang_code, param, code_desc, template_tag, not_real_lang)
return require("Module:languages/error")(lang_code, param, code_desc, template_tag, not_real_lang)

Latest revision as of 17:41, 28 April 2024

The following documentation is located at Module:languages/documentation. [edit]
Useful links: subpage list • links • transclusions • testcases • sandbox

This module is used to retrieve and manage the languages that can have Wiktionary entries, and the information associated with them. See Wiktionary:Languages for more information.

For the languages and language varieties that may be used in etymologies, see Module:etymology languages. For language families, which sometimes also appear in etymologies, see Module:families.

This module provides access to other modules. To access the information from within a template, see Module:languages/templates.

The information itself is stored in the various data modules that are subpages of this module. These modules should not be used directly by any other module, the data should only be accessed through the functions provided by this module.

Data submodules:

Extra data submodules (for less frequently used data):

Finding and retrieving languages

The module exports a number of functions that are used to find languages.

export.getDataModuleName

function export.getDataModuleName(code)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

export.getExtraDataModuleName

function export.getExtraDataModuleName(code)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

export.makeObject

function export.makeObject(code, data, useRequire, dontCanonicalizeAliases)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

export.getByCode

function export.getByCode(code, paramForError, allowEtymLang, allowFamily, useRequire)

Finds the language whose code matches the one provided. If it exists, it returns a Language object representing the language. Otherwise, it returns nil, unless paramForError is given, in which case an error is generated. If paramForError is true, a generic error message mentioning the bad code is generated; otherwise paramForError should be a string or number specifying the parameter that the code came from, and this parameter will be mentioned in the error message along with the bad code. If allowEtymLang is specified, etymology-only language codes are allowed and looked up along with normal language codes. If allowFamily is specified, language family codes are allowed and looked up along with normal language codes.

export.getByCanonicalName

function export.getByCanonicalName(name, errorIfInvalid, allowEtymLang, allowFamily, useRequire)

Finds the language whose canonical name (the name used to represent that language on Wiktionary) or other name matches the one provided. If it exists, it returns a Language object representing the language. Otherwise, it returns nil, unless paramForError is given, in which case an error is generated. If allowEtymLang is specified, etymology-only language codes are allowed and looked up along with normal language codes. If allowFamily is specified, language family codes are allowed and looked up along with normal language codes. The canonical name of languages should always be unique (it is an error for two languages on Wiktionary to share the same canonical name), so this is guaranteed to give at most one result. This function is powered by Module:languages/canonical names, which contains a pre-generated mapping of full-language canonical names to codes. It is generated by going through the Category:Language data modules for full languages. When allowEtymLang is specified for the above function, Module:etymology languages/canonical names may also be used, and when allowFamily is specified for the above function, Module:families/canonical names may also be used.

export.addDefaultTypes

function export.addDefaultTypes(data, regular, ...)

Used by Module:languages/data/2 (et al.) to add default types to the entities returned.

export.finalizeLanguageData

function export.finalizeLanguageData(data)

Used by Module:languages/data/2 (et al.) and Module:etymology languages/data to finalize language-related data into the format that is actually returned.

export.finalizeEtymologyData

function export.finalizeEtymologyData(data)

Used by Module:etymology languages/data and Module:families/data/etymology to finalize etymology-related data into the format that is actually returned.

export.err

function export.err(lang_code, param, code_desc, template_tag, not_real_lang)

For backwards compatibility only; modules should require the error themselves.

Language objects

A Language object is returned from one of the functions above. It is a Lua representation of a language and the data associated with it. It has a number of methods that can be called on it, using the : syntax. For example: 

local m_languages = require("Module:languages")
local lang = m_languages.getByCode("fr")
local name = lang:getCanonicalName()
-- "name" will now be "French"

Language:loadData

function Language:loadData(modname)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getCode

function Language:getCode()

Returns the language code of the language. Example: "fr" for French.

Language:getCanonicalName

function Language:getCanonicalName()

Returns the canonical name of the language. This is the name used to represent that language on Wiktionary, and is guaranteed to be unique to that language alone. Example: "French" for French.

Language:getDisplayForm

function Language:getDisplayForm()

Return the display form of the language. The display form of a language, family or script is the form it takes when appearing as the source in categories such as English terms derived from source or English given names from source, and is also the displayed text in makeCategoryLink() links. For full and etymology-only languages, this is the same as the canonical name, but for families, it reads "name languages" (e.g. "Indo-Iranian languages"), and for scripts, it reads "name script" (e.g. "Arabic script").

Language:getOtherNames

function Language:getOtherNames(onlyOtherNames)

Returns a table of the "other names" that the language is known by, excluding the canonical name. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {"Manx Gaelic", "Northern Manx", "Southern Manx"} for Manx. If onlyOtherNames is given and is non-nil, only names explicitly listed in the otherNames field are returned; otherwise, names listed under otherNames, aliases and varieties are combined together and returned. For example, for Manx, Manx Gaelic is listed as an alias, while Northern Manx and Southern Manx are listed as varieties. It should be noted that the otherNames field itself is deprecated, and entries listed there should eventually be moved to either aliases or varieties.

Language:getAliases

function Language:getAliases()

Returns a table of the aliases that the language is known by, excluding the canonical name. Aliases are synonyms for the language in question. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {"High German", "New High German", "Deutsch"} for German.

Language:getVarieties

function Language:getVarieties(flatten)

Return a table of the known subvarieties of a given language, excluding subvarieties that have been given explicit etymology-only language codes. The names are not guaranteed to be unique, in that sometimes a given name refers to a subvariety of more than one language. Example: {"Southern Aymara", "Central Aymara"} for Aymara. Note that the returned value can have nested tables in it, when a subvariety goes by more than one name. Example: {"North Azerbaijani", "South Azerbaijani", {"Afshar", "Afshari", "Afshar Azerbaijani", "Afchar"}, {"Qashqa'i", "Qashqai", "Kashkay"}, "Sonqor"} for Azerbaijani. Here, for example, Afshar, Afshari, Afshar Azerbaijani and Afchar all refer to the same subvariety, whose preferred name is Afshar (the one listed first). To avoid a return value with nested tables in it, specify a non-nil value for the flatten parameter; in that case, the return value would be {"North Azerbaijani", "South Azerbaijani", "Afshar", "Afshari", "Afshar Azerbaijani", "Afchar", "Qashqa'i", "Qashqai", "Kashkay", "Sonqor"}.

Language:getTypes

function Language:getTypes()

Returns a table of types as a lookup table (with the types as keys).

The possible types are

  • language: This is a language, either full or etymology-only.
  • full: This is a "full" (not etymology-only) language, i.e. the union of regular, reconstructed and appendix-constructed. Note that the types full and etymology-only also exist for families, so if you want to check specifically for a full language and you have an object that might be a family, you should use hasType("language", "full") and not simply hasType("full").
  • etymology-only: This is an etymology-only (not full) language, whose parent is another etymology-only language or a full language. Note that the types full and etymology-only also exist for families, so if you want to check specifically for an etymology-only language and you have an object that might be a family, you should use hasType("language", "etymology-only") and not simply hasType("etymology-only").
  • regular: This indicates a full language that is attested according to WT:CFI and therefore permitted in the main namespace. There may also be reconstructed terms for the language, which are placed in the Reconstruction namespace and must be prefixed with * to indicate a reconstruction. Most full languages are natural (not constructed) languages, but a few constructed languages (e.g. Esperanto and Volapük, among others) are also allowed in the mainspace and considered regular languages.
  • reconstructed: This language is not attested according to WT:CFI, and therefore is allowed only in the Reconstruction namespace. All terms in this language are reconstructed, and must be prefixed with *. Languages such as Proto-Indo-European and Proto-Germanic are in this category.
  • appendix-constructed: This language is attested but does not meet the additional requirements set out for constructed languages (WT:CFI#Constructed languages). Its entries must therefore be in the Appendix namespace, but they are not reconstructed and therefore should not have * prefixed in links.

Language:hasType

function Language:hasType(...)

Given a list of types as strings, returns true if the language has all of them.

Language:getWikimediaLanguages

function Language:getWikimediaLanguages()

Returns a table containing WikimediaLanguage objects (see Module:wikimedia languages), which represent languages and their codes as they are used in Wikimedia projects for interwiki linking and such. More than one object may be returned, as a single Wiktionary language may correspond to multiple Wikimedia languages. For example, Wiktionary's single code sh (Serbo-Croatian) maps to four Wikimedia codes: sh (Serbo-Croatian), bs (Bosnian), hr (Croatian) and sr (Serbian). The code for the Wikimedia language is retrieved from the wikimedia_codes property in the data modules. If that property is not present, the code of the current language is used. If none of the available codes is actually a valid Wikimedia code, an empty table is returned.

Language:getWikimediaLanguageCodes

function Language:getWikimediaLanguageCodes()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getWikipediaArticle

function Language:getWikipediaArticle(noCategoryFallback, project)

Returns the name of the Wikipedia article for the language. project specifies the language and project to retrieve the article from, defaulting to "enwiki" for the English Wikipedia. Normally if specified it should be the project code for a specific-language Wikipedia e.g. "zhwiki" for the Chinese Wikipedia, but it can be any project, including non-Wikipedia ones. If the project is the English Wikipedia and the property wikipedia_article is present in the data module it will be used first. In all other cases, a sitelink will be generated from :getWikidataItem (if set). The resulting value (or lack of value) is cached so that subsequent calls are fast. If no value could be determined, and noCategoryFallback is false, :getCategoryName is used as fallback; otherwise, nil is returned. Note that if noCategoryFallback is nil or omitted, it defaults to false if the project is the English Wikipedia, otherwise to true. In other words, under normal circumstances, if the English Wikipedia article couldn't be retrieved, the return value will fall back to a link to the language's category, but this won't normally happen for any other project.

Language:makeWikipediaLink

function Language:makeWikipediaLink()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getWikidataItem

function Language:getWikidataItem()

Returns the Wikidata item id for the language or nil. This corresponds to the the second field in the data modules.

Language:getScripts

function Language:getScripts()

Returns a table of Script objects for all scripts that the language is written in. See Module:scripts.

Language:getScriptCodes

function Language:getScriptCodes()

Returns the table of script codes in the language's data file.

Language:findBestScript

function Language:findBestScript(text, forceDetect)

Given some text, this function iterates through the scripts of a given language and tries to find the script that best matches the text. It returns a Script object representing the script. If no match is found at all, it returns the None script object.

Language:getFamily

function Language:getFamily()

Returns a Family object for the language family that the language belongs to. See Module:families.

Language:getFamilyCode

function Language:getFamilyCode()

Returns the family code in the language's data file.

Language:getFamilyName

function Language:getFamilyName()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:inFamily

function Language:inFamily(...)

Check whether the language belongs to family (which can be a family code or object). A list of objects can be given in place of family; in that case, return true if the language belongs to any of the specified families. Note that some languages (in particular, certain creoles) can have multiple immediate ancestors potentially belonging to different families; in that case, return true if the language belongs to any of the specified families.

Language:getParent

function Language:getParent()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getParentCode

function Language:getParentCode()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getParentName

function Language:getParentName()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getParentChain

function Language:getParentChain()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:hasParent

function Language:hasParent(...)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getFull

function Language:getFull()

If the language is etymology-only, this iterates through parents until a full language or family is found, and the corresponding object is returned. If the language is a full language, then it simply returns itself.

Language:getFullCode

function Language:getFullCode()

If the language is an etymology-only language, this iterates through parents until a full language or family is found, and the corresponding code is returned. If the language is a full language, then it simply returns the language code.

Language:getFullName

function Language:getFullName()

If the language is an etymology-only language, this iterates through parents until a full language or family is found, and the corresponding canonical name is returned. If the language is a full language, then it simply returns the canonical name of the language.

Language:getAncestors

function Language:getAncestors()

Returns a table of Language objects for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.

Language:getAncestorCodes

function Language:getAncestorCodes()

Returns a table of Language codes for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.

Language:hasAncestor

function Language:hasAncestor(...)

Given a list of language objects or codes, returns true if at least one of them is an ancestor. This includes any etymology-only children of that ancestor. If the language's ancestor(s) are etymology-only languages, it will also return true for those language parent(s) (e.g. if Vulgar Latin is the ancestor, it will also return true for its parent, Latin). However, a parent is excluded from this if the ancestor is also ancestral to that parent (e.g. if Classical Persian is the ancestor, Persian would return false, because Classical Persian is also ancestral to Persian).

Language:getAncestorChain

function Language:getAncestorChain()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getDescendants

function Language:getDescendants()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getDescendantCodes

function Language:getDescendantCodes()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getDescendantNames

function Language:getDescendantNames()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:hasDescendant

function Language:hasDescendant(...)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getChildren

function Language:getChildren()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getChildrenCodes

function Language:getChildrenCodes()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getChildrenNames

function Language:getChildrenNames()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:hasChild

function Language:hasChild(...)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getCategoryName

function Language:getCategoryName(nocap)

Returns the name of the main category of that language. Example: "French language" for French, whose category is at Category:French language. Unless optional argument nocap is given, the language name at the beginning of the returned value will be capitalized. This capitalization is correct for category names, but not if the language name is lowercase and the returned value of this function is used in the middle of a sentence.

Language:makeCategoryLink

function Language:makeCategoryLink()

Creates a link to the category; the link text is the canonical name.

Language:getStandardCharacters

function Language:getStandardCharacters(sc)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:makeEntryName

function Language:makeEntryName(text, sc)

Make the entry name (i.e. the correct page name).

Language:generateForms

function Language:generateForms(text, sc)

Generates alternative forms using a specified method, and returns them as a table. If no method is specified, returns a table containing only the input term.

Language:makeSortKey

function Language:makeSortKey(text, sc)

Creates a sort key for the given entry name, following the rules appropriate for the language. This removes diacritical marks from the entry name if they are not considered significant for sorting, and may perform some other changes. Any initial hyphen is also removed, and anything parentheses is removed as well. The sort_key setting for each language in the data modules defines the replacements made by this function, or it gives the name of the module that takes the entry name and returns a sortkey.

Language:makeDisplayText

function Language:makeDisplayText(text, sc, keepPrefixes)

Make the display text (i.e. what is displayed on the page).

Language:transliterate

function Language:transliterate(text, sc, module_override)

Transliterates the text from the given script into the Latin script (see Wiktionary:Transliteration and romanization). The language must have the translit property for this to work; if it is not present, nil is returned. Returns three values:

  1. The transliteration.
  2. A boolean which indicates whether the transliteration failed for an unexpected reason. If false, then the transliteration either succeeded, or the module is returning nothing in a controlled way (e.g. the input was "-"). Generally, this means that no maintenance action is required. If true, then the transliteration is nil because either the input or output was defective in some way (e.g. Module:ar-translit will not transliterate non-vocalised inputs, and this module will fail partially-completed transliterations in all languages). Note that this value can be manually set by the transliteration module, so make sure to cross-check to ensure it is accurate.
  3. A table of categories selected by the transliteration module, which should be in the format expected by format_categories in Module:utilities. The sc parameter is handled by the transliteration module, and how it is handled is specific to that module. Some transliteration modules may tolerate nil as the script, others require it to be one of the possible scripts that the module can transliterate, and will show an error if it's not one of them. For this reason, the sc parameter should always be provided when writing non-language-specific code. The module_override parameter is used to override the default module that is used to provide the transliteration. This is useful in cases where you need to demonstrate a particular module in use, but there is no default module yet, or you want to demonstrate an alternative version of a transliteration module before making it official. It should not be used in real modules or templates, only for testing. All uses of this parameter are tracked by Wiktionary:Tracking/module_override. Known bugs:
  • This function assumes tr(s1) .. tr(s2) == tr(s1 .. s2). When this assertion fails, wikitext markups like ''' can cause wrong transliterations.
  • HTML entities like &apos;, often used to escape wikitext markups, do not work.

Language:overrideManualTranslit

function Language:overrideManualTranslit()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:hasTranslit

function Language:hasTranslit()

Returns true if the language has a transliteration module, or false if it doesn't.

Language:link_tr

function Language:link_tr()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:hasDottedDotlessI

function Language:hasDottedDotlessI()

Returns true if the language uses the letters I/ı and İ/i, or false if it doesn't.

Language:toJSON

function Language:toJSON(returnTable)

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Language:getRawData

function Language:getRawData()

This function is not for use in entries or other content pages. Returns a blob of data about the language. The format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes. -- Do NOT use these methods! -- All uses should be pre-approved on the talk page!

Language:getRawExtraData

function Language:getRawExtraData()

This function is not for use in entries or other content pages. Returns a blob of data about the language that contains the "extra data". Much like with getRawData, the format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.

Language:loadInExtraData

function Language:loadInExtraData()

This function lacks documentation. Please add a description of its usages, inputs and outputs, or its difference from similar functions, or make it local to remove it from the function list.

Error function

See Module:languages/error.

Subpages

See also

local require = require

local m_str_utils = require("Module:string utilities")
local m_table = require("Module:table")
local mw = mw
local string = string
local table = table
local ustring = mw.ustring

local check_object = require("Module:utilities").check_object
local concat = table.concat
local decode_entities = m_str_utils.decode_entities
local decode_uri = m_str_utils.decode_uri
local gmatch = string.gmatch
local gsub = string.gsub
local insert = table.insert
local ipairs = ipairs
local load_data = mw.loadData
local match = string.match
local next = next
local pairs = pairs
local remove = table.remove
local remove_duplicates = m_table.removeDuplicates
local select = select
local setmetatable = setmetatable
local shallowcopy = m_table.shallowcopy
local split = m_str_utils.split
local type = type
local ugsub = ustring.gsub
local ulen = m_str_utils.len
local umatch = ustring.match

-- Loaded as needed by findBestScript.
local Hans_chars
local Hant_chars

local export = {}

--[=[
This module implements fetching of language-specific information and processing text in a given language.

There are two types of languages: full languages and etymology-only languages. The essential difference is that only
full languages appear in L2 headings in vocabulary entries, and hence categories like [[:Category:French nouns]] exist
only for full languages. Etymology-only languages have either a full language or another etymology-only language as
their parent (in the parent-child inheritance sense), and for etymology-only languages with another etymology-only
language as their parent, a full language can always be derived by following the parent links upwards. For example,
"Canadian French", code 'fr-CA', is an etymology-only language whose parent is the full language "French", code 'fr'.
An example of an etymology-only language with another etymology-only parent is "Northumbrian Old English", code
'ang-nor', which has "Anglian Old English", code 'ang-ang' as its parent; this is an etymology-only language whose
parent is "Old English", code "ang", which is a full language. (This is because Northumbrian Old English is considered
a variety of Anglian Old English.) Sometimes the parent is the "Undetermined" language, code 'und'; this is the case,
for example, for "substrate" languages such as "Pre-Greek", code 'qsb-grc', and "the BMAC substrate", code 'qsb-bma'.

It is important to distinguish language ''parents'' from language ''ancestors''. The parent-child relationship is one
of containment, i.e. if X is a child of Y, X is considered a variety of Y. On the other hand, the ancestor-descendant
relationship is one of descent in time. For example, "Classical Latin", code 'la-cla', and "Late Latin", code 'la-lat',
are both etymology-only languages with "Latin", code 'la', as their parents, because both of the former are varieties
of Latin. However, Late Latin does *NOT* have Classical Latin as its parent because Late Latin is *not* a variety of
Classical Latin; rather, it is a descendant. There is in fact a separate 'ancestors' field that is used to express the
ancestor-descendant relationship, and Late Latin's ancestor is given as Classical Latin. It is also important to note
that sometimes an etymology-only language is actually the conceptual ancestor of its parent language. This happens,
for example, with "Old Italian" (code 'roa-oit'), which is an etymology-only variant of full language "Italian" (code
'it'), and with "Old Latin" (code 'itc-ola'), which is an etymology-only variant of Latin. In both cases, the full
language has the etymology-only variant listed as an ancestor. This allows a Latin term to inherit from Old Latin
using the {{tl|inh}} template (where in this template, "inheritance" refers to ancestral inheritance, i.e. inheritance
in time, rather than in the parent-child sense); likewise for Italian and Old Italian.

Full languages come in three subtypes:
* {regular}: This indicates a full language that is attested according to [[WT:CFI]] and therefore permitted in the
			 main namespace. There may also be reconstructed terms for the language, which are placed in the
			 {Reconstruction} namespace and must be prefixed with * to indicate a reconstruction. Most full languages
			 are natural (not constructed) languages, but a few constructed languages (e.g. Esperanto and Volapük,
			 among others) are also allowed in the mainspace and considered regular languages.
* {reconstructed}: This language is not attested according to [[WT:CFI]], and therefore is allowed only in the
				   {Reconstruction} namespace. All terms in this language are reconstructed, and must be prefixed with
				   *. Languages such as Proto-Indo-European and Proto-Germanic are in this category.
* {appendix-constructed}: This language is attested but does not meet the additional requirements set out for
						  constructed languages ([[WT:CFI#Constructed languages]]). Its entries must therefore be in
						  the Appendix namespace, but they are not reconstructed and therefore should not have *
						  prefixed in links. Most constructed languages are of this subtype.

Both full languages and etymology-only languages have a {Language} object associated with them, which is fetched using
the {getByCode} function in [[Module:languages]] to convert a language code to a {Language} object. Depending on the
options supplied to this function, etymology-only languages may or may not be accepted, and family codes may be
accepted (returning a {Family} object as described in [[Module:families]]). There are also separate {getByCanonicalName}
functions in [[Module:languages]] and [[Module:etymology languages]] to convert a language's canonical name to a
{Language} object (depending on whether the canonical name refers to a full or etymology-only language).

Textual strings belonging to a given language come in several different ''text variants'':
# The ''input text'' is what the user supplies in wikitext, in the parameters to {{tl|m}}, {{tl|l}}, {{tl|ux}},
  {{tl|t}}, {{tl|lang}} and the like.
# The ''display text'' is the text in the form as it will be displayed to the user. This can include accent marks that
  are stripped to form the entry text (see below), as well as embedded bracketed links that are variously processed
  further. The display text is generated from the input text by applying language-specific transformations; for most
  languages, there will be no such transformations. Examples of transformations are bad-character replacements for
  certain languages (e.g. replacing 'l' or '1' to [[palochka]] in certain languages in Cyrillic); and for Thai and
  Khmer, converting space-separated words to bracketed words and resolving respelling substitutions such as [กรีน/กฺรีน],
  which indicate how to transliterate given words.
# The ''entry text'' is the text in the form used to generate a link to a Wiktionary entry. This is usually generated
  from the display text by stripping certain sorts of diacritics on a per-language basis, and sometimes doing other
  transformations. The concept of ''entry text'' only really makes sense for text that does not contain embedded links,
  meaning that display text containing embedded links will need to have the links individually processed to get
  per-link entry text in order to generate the resolved display text (see below).
# The ''resolved display text'' is the result of resolving embedded links in the display text (e.g. converting them to
  two-part links where the first part has entry-text transformations applied, and adding appropriate language-specific
  fragments) and adding appropriate language and script tagging. This text can be passed directly to MediaWiki for
  display.
# The ''source translit text'' is the text as supplied to the language-specific {transliterate()} method. The form of
  the source translit text may need to be language-specific, e.g Thai and Khmer will need the full unprocessed input
  text, whereas other languages may need to work off the display text. [FIXME: It's still unclear to me how embedded
  bracketed links are handled in the existing code.] In general, embedded links need to be removed (i.e. converted to
  their "bare display" form by taking the right part of two-part links and removing double brackets), but when this
  happens is unclear to me [FIXME]. Some languages have a chop-up-and-paste-together scheme that sends parts of the
  text through the transliterate mechanism, and for others (those listed in {contiguous_substition} in
  [[Module:languages/data]]) they receive the full input text, but preprocessed in certain ways. (The wisdom of this is
  still unclear to me.)
# The ''transliterated text'' (or ''transliteration'') is the result of transliterating the source translit text.
  Unlike for all the other text variants except the transcribed text, it is always in the Latin script.
# The ''transcribed text'' (or ''transcription'') is the result of transcribing the source translit text, where
  "transcription" here means a close approximation to the phonetic form of the language in languages (e.g. Akkadian,
  Sumerian, Ancient Egyptian, maybe Tibetan) that have a wide difference between the written letters and spoken form.
  Unlike for all the other text variants other than the transliterated text, it is always in the Latin script.
  Currently, the transcribed text is always supplied manually be the user; there is no such thing as a
  {lua|transcribe()} method on language objects.
# The ''sort key'' is the text used in sort keys for determining the placing of pages in categories they belong to. The
  sort key is generated from the pagename or a specified ''sort base'' by lowercasing, doing language-specific
  transformations and then uppercasing the result. If the sort base is supplied and is generated from input text, it
  needs to be converted to display text, have embedded links removed (i.e. resolving them to their right side if they
  are two-part links) and have entry text transformations applied.
# There are other text variants that occur in usexes (specifically, there are normalized variants of several of the
  above text variants), but we can skip them for now.

The following methods exist on {Language} objects to convert between different text variants:
# {makeDisplayText}: This converts input text to display text.
# {lua|makeEntryName}: This converts input or display text to entry text. [FIXME: This needs some rethinking. In
  particular, {lua|makeEntryName} is sometimes called on display text (in some paths inside of [[Module:links]]) and
  sometimes called on input text (in other paths inside of [[Module:links]], and usually from other modules). We need
  to make sure we don't try to convert input text to display text twice, but at the same time we need to support
  calling it directly on input text since so many modules do this. This means we need to add a parameter indicating
  whether the passed-in text is input or display text; if that former, we call {lua|makeDisplayText} ourselves.]
# {lua|transliterate}: This appears to convert input text with embedded brackets removed into a transliteration.
  [FIXME: This needs some rethinking. In particular, it calls {lua|processDisplayText} on its input, which won't work
  for Thai and Khmer, so we may need language-specific flags indicating whether to pass the input text directly to the
  language transliterate method. In addition, I'm not sure how embedded links are handled in the existing translit code;
  a lot of callers remove the links themselves before calling {lua|transliterate()}, which I assume is wrong.]
# {lua|makeSortKey}: This converts entry text (?) to a sort key. [FIXME: Clarify this.]
]=]

local function track(page)
	require("Module:debug/track")("languages/" .. page)
	return true
end

local function normalize_code(code)
	return load_data("Module:languages/data").aliases[code] or code
end

-- Convert risky characters to HTML entities, which minimizes interference once returned (e.g. for "sms:a", "<!-- -->" etc.).
local function escape_risky_characters(text)
	local encode_entities = require("Module:string/encode entities")
	-- Spacing characters in isolation generally need to be escaped in order to be properly processed by the MediaWiki software.
	if umatch(text, "^%s*$") then
		return encode_entities(text, text)
	else
		return encode_entities(text, "!#%&*+/:;<=>?@[\\]_{|}")
	end
end

-- Temporarily convert various formatting characters to PUA to prevent them from being disrupted by the substitution process.
local function doTempSubstitutions(text, subbedChars, keepCarets, noTrim)
	-- Clone so that we don't insert any extra patterns into the table in package.loaded. For some reason, using require seems to keep memory use down; probably because the table is always cloned.
	local patterns = shallowcopy(require("Module:languages/data/patterns"))
	if keepCarets then
		insert(patterns, "((\\+)%^)")
		insert(patterns, "((%^))")
	end
	-- Ensure any whitespace at the beginning and end is temp substituted, to prevent it from being accidentally trimmed. We only want to trim any final spaces added during the substitution process (e.g. by a module), which means we only do this during the first round of temp substitutions.
	if not noTrim then
		insert(patterns, "^([\128-\191\244]*(%s+))")
		insert(patterns, "((%s+)[\128-\191\244]*)$")
	end
	-- Pre-substitution, of "[[" and "]]", which makes pattern matching more accurate.
	text = text
		:gsub("%f[%[]%[%[", "\1")
		:gsub("%f[%]]%]%]", "\2")
	local i, pe = #subbedChars, require("Module:string utilities").pattern_escape
	for _, pattern in ipairs(patterns) do
		-- Patterns ending in \0 stand are for things like "[[" or "]]"), so the inserted PUA are treated as breaks between terms by modules that scrape info from pages.
		local term_divider
		pattern = pattern:gsub("%z$", function(divider)
			term_divider = divider == "\0"
			return ""
		end)
		text = text:gsub(pattern, function(...)
			local m = {...}
			local m1New = m[1]
			for k = 2, #m do
				local n = i + k - 1
				subbedChars[n] = m[k]
				local byte2 = math.floor(n / 4096) % 64 + (term_divider and 128 or 136)
				local byte3 = math.floor(n / 64) % 64 + 128
				local byte4 = n % 64 + 128
				m1New = m1New:gsub(pe(m[k]), "\244" .. string.char(byte2) .. string.char(byte3) .. string.char(byte4), 1)
			end
			i = i + #m - 1
			return m1New
		end)
	end
	text = text
		:gsub("\1", "%[%[")
		:gsub("\2", "%]%]")
	return text, subbedChars
end

-- Reinsert any formatting that was temporarily substituted.
local function undoTempSubstitutions(text, subbedChars)
	local pe = require("Module:string utilities").pattern_escape
	for i = 1, #subbedChars do
		local byte2 = math.floor(i / 4096) % 64 + 128
		local byte3 = math.floor(i / 64) % 64 + 128
		local byte4 = i % 64 + 128
		text = text:gsub("\244[" .. string.char(byte2) .. string.char(byte2+8) .. "]" .. string.char(byte3) .. string.char(byte4), pe(subbedChars[i]))
	end
	text = text
		:gsub("\1", "%[%[")
		:gsub("\2", "%]%]")
	return text
end

-- Check if the raw text is an unsupported title, and if so return that. Otherwise, remove HTML entities. We do the pre-conversion to avoid loading the unsupported title list unnecessarily.
local function checkNoEntities(self, text)
	local textNoEnc = decode_entities(text)
	if textNoEnc ~= text and self:loadData("Module:links/data").unsupported_titles[text] then
		return text
	else
		return textNoEnc
	end
end

-- If no script object is provided (or if it's invalid or None), get one.
local function checkScript(text, self, sc)
	if not check_object("script", true, sc) or sc:getCode() == "None" then
		return self:findBestScript(text)
	else
		return sc
	end
end

local function normalize(text, sc)
	text = sc:fixDiscouragedSequences(text)
	return sc:toFixedNFD(text)
end

-- Split the text into sections, based on the presence of temporarily substituted formatting characters, then iterate over each one to apply substitutions. This avoids putting PUA characters through language-specific modules, which may be unequipped for them.
local function iterateSectionSubstitutions(text, subbedChars, keepCarets, self, sc, substitution_data, function_name)
	local pe = require("Module:string utilities").pattern_escape
	local fail, cats, sections = nil, {}
	-- See [[Module:languages/data]].
	if not text:find("\244") or self:loadData("Module:languages/data").contiguous_substitution[self._code] then
		sections = {text}
	else
		sections = split(text, "\244[\128-\143][\128-\191]*", true)
	end
	for _, section in ipairs(sections) do
		-- Don't bother processing empty strings or whitespace (which may also not be handled well by dedicated modules).
		if section:gsub("%s+", "") ~= "" then
			local sub, sub_fail, sub_cats = require("Module:languages/doSubstitutions")(section, self, sc, substitution_data, function_name)
			-- Second round of temporary substitutions, in case any formatting was added by the main substitution process. However, don't do this if the section contains formatting already (as it would have had to have been escaped to reach this stage, and therefore should be given as raw text).
			if sub and subbedChars then
				local noSub
				for _, pattern in ipairs(require("Module:languages/data/patterns")) do
					if section:match(pattern .. "%z?") then
						noSub = true
					end
				end
				if not noSub then
					sub, subbedChars = doTempSubstitutions(sub, subbedChars, keepCarets, true)
				end
			end
			if (not sub) or sub_fail then
				text = sub
				fail = sub_fail
				cats = sub_cats or {}
				break
			end
			text = sub and text:gsub(pe(section), pe(sub), 1) or text
			if type(sub_cats) == "table" then
				for _, cat in ipairs(sub_cats) do
					insert(cats, cat)
				end
			end
		end
	end

	-- Trim, unless there are only spacing characters, while ignoring any final formatting characters.
	text = text and text
		:gsub("^([\128-\191\244]*)%s+(%S)", "%1%2")
		:gsub("(%S)%s+([\128-\191\244]*)$", "%1%2")

	-- Remove duplicate categories.
	if #cats > 1 then
		cats = remove_duplicates(cats)
	end

	return text, fail, cats, subbedChars
end

-- Process carets (and any escapes). Default to simple removal, if no pattern/replacement is given.
local function processCarets(text, pattern, repl)
	local rep
	repeat
		text, rep = text:gsub("\\\\(\\*^)", "\3%1")
	until rep == 0
	return text
		:gsub("\\^", "\4")
		:gsub(pattern or "%^", repl or "")
		:gsub("\3", "\\")
		:gsub("\4", "^")
end

-- Remove carets if they are used to capitalize parts of transliterations (unless they have been escaped).
local function removeCarets(text, sc)
	if not sc:hasCapitalization() and sc:isTransliterated() and text:match("%^") then
		return processCarets(text)
	else
		return text
	end
end

local Language = {}

function Language:loadData(modname)
	return (self._useRequire and require or mw.loadData)(modname)
end

--[==[Returns the language code of the language. Example: {{code|lua|"fr"}} for French.]==]
function Language:getCode()
	return self._code
end

--[==[Returns the canonical name of the language. This is the name used to represent that language on Wiktionary, and is guaranteed to be unique to that language alone. Example: {{code|lua|"French"}} for French.]==]
function Language:getCanonicalName()
	local name = self._name
	if name == nil then
		name = self._rawData[1]
		self._name = name
	end
	return name
end

--[==[
Return the display form of the language. The display form of a language, family or script is the form it takes when
appearing as the <code><var>source</var></code> in categories such as <code>English terms derived from
<var>source</var></code> or <code>English given names from <var>source</var></code>, and is also the displayed text
in {makeCategoryLink()} links. For full and etymology-only languages, this is the same as the canonical name, but
for families, it reads <code>"<var>name</var> languages"</code> (e.g. {"Indo-Iranian languages"}), and for scripts,
it reads <code>"<var>name</var> script"</code> (e.g. {"Arabic script"}).
]==]
function Language:getDisplayForm()
	local form = self._displayForm
	if form == nil then
		form = self:getCanonicalName()
		-- Add article and " substrate" to substrates that lack them.
		if self:getFamilyCode() == "qfa-sub" then
			if not (match(form, "^[Tt]he ") or match(form, "^[Aa] ")) then
				form = "a " .. form
			end
			if not match(form, "[Ss]ubstrate") then
				form = form .. " substrate"
			end
		end
		self._displayForm = form
	end
	return form
end

--[==[Returns a table of the "other names" that the language is known by, excluding the canonical name. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {{code|lua|{"Manx Gaelic", "Northern Manx", "Southern Manx"} }} for [[:Category:Manx language|Manx]]. If <code>onlyOtherNames</code> is given and is non-{{code|lua|nil}}, only names explicitly listed in the <code>otherNames</code> field are returned; otherwise, names listed under <code>otherNames</code>, <code>aliases</code> and <code>varieties</code> are combined together and returned. For example, for Manx, Manx Gaelic is listed as an alias, while Northern Manx and Southern Manx are listed as varieties. It should be noted that the <code>otherNames</code> field itself is deprecated, and entries listed there should eventually be moved to either <code>aliases</code> or <code>varieties</code>.]==]
function Language:getOtherNames(onlyOtherNames)
	if #self._stack == 1 then
		self:loadInExtraData()
	end
	return require("Module:language-like").getOtherNames(self, onlyOtherNames)
end

--[==[Returns a table of the aliases that the language is known by, excluding the canonical name. Aliases are synonyms for the language in question. The names are not guaranteed to be unique, in that sometimes more than one language is known by the same name. Example: {{code|lua|{"High German", "New High German", "Deutsch"} }} for [[:Category:German language|German]].]==]
function Language:getAliases()
	if #self._stack == 1 then
		self:loadInExtraData()
	end
	return self._rawData.aliases or (self._extraData and self._extraData.aliases) or {}
end

--[==[
Return a table of the known subvarieties of a given language, excluding subvarieties that have been given
explicit etymology-only language codes. The names are not guaranteed to be unique, in that sometimes a given name
refers to a subvariety of more than one language. Example: {{code|lua|{"Southern Aymara", "Central Aymara"} }} for
[[:Category:Aymara language|Aymara]]. Note that the returned value can have nested tables in it, when a subvariety
goes by more than one name. Example: {{code|lua|{"North Azerbaijani", "South Azerbaijani", {"Afshar", "Afshari",
"Afshar Azerbaijani", "Afchar"}, {"Qashqa'i", "Qashqai", "Kashkay"}, "Sonqor"} }} for
[[:Category:Azerbaijani language|Azerbaijani]]. Here, for example, Afshar, Afshari, Afshar Azerbaijani and Afchar
all refer to the same subvariety, whose preferred name is Afshar (the one listed first). To avoid a return value
with nested tables in it, specify a non-{{code|lua|nil}} value for the <code>flatten</code> parameter; in that case,
the return value would be {{code|lua|{"North Azerbaijani", "South Azerbaijani", "Afshar", "Afshari",
"Afshar Azerbaijani", "Afchar", "Qashqa'i", "Qashqai", "Kashkay", "Sonqor"} }}.
]==]
function Language:getVarieties(flatten)
	if #self._stack == 1 then
		self:loadInExtraData()
	end
	return require("Module:language-like").getVarieties(self, flatten)
end

--[==[Returns a table of types as a lookup table (with the types as keys). 

The possible types are
* {language}: This is a language, either full or etymology-only.
* {full}: This is a "full" (not etymology-only) language, i.e. the union of {regular}, {reconstructed} and
		  {appendix-constructed}. Note that the types {full} and {etymology-only} also exist for families, so if you
		  want to check specifically for a full language and you have an object that might be a family, you should
		  use {{lua|hasType("language", "full")}} and not simply {{lua|hasType("full")}}.
* {etymology-only}: This is an etymology-only (not full) language, whose parent is another etymology-only
					language or a full language. Note that the types {full} and {etymology-only} also exist for
					families, so if you want to check specifically for an etymology-only language and you have an
					object that might be a family, you should use {{lua|hasType("language", "etymology-only")}}
					and not simply {{lua|hasType("etymology-only")}}.
* {regular}: This indicates a full language that is attested according to [[WT:CFI]] and therefore permitted
			 in the main namespace. There may also be reconstructed terms for the language, which are placed in
			 the {Reconstruction} namespace and must be prefixed with * to indicate a reconstruction. Most full
			 languages are natural (not constructed) languages, but a few constructed languages (e.g. Esperanto
			 and Volapük, among others) are also allowed in the mainspace and considered regular languages.
* {reconstructed}: This language is not attested according to [[WT:CFI]], and therefore is allowed only in the
				   {Reconstruction} namespace. All terms in this language are reconstructed, and must be prefixed
				   with *. Languages such as Proto-Indo-European and Proto-Germanic are in this category.
* {appendix-constructed}: This language is attested but does not meet the additional requirements set out for
						  constructed languages ([[WT:CFI#Constructed languages]]). Its entries must therefore
						  be in the Appendix namespace, but they are not reconstructed and therefore should
						  not have * prefixed in links.
]==]
function Language:getTypes()
	local types = self._types
	if types == nil then
		types = {language = true}
		if self:getFullCode() == self._code then
			types.full = true
		else
			types["etymology-only"] = true
		end
		for t in gmatch(self._rawData.type, "[^,]+") do
			types[t] = true
		end
		self._types = types
	end
	return types
end

--[==[Given a list of types as strings, returns true if the language has all of them.]==]
function Language:hasType(...)
	local args, types = {...}, self:getTypes()
	for i = 1, #args do
		if not types[args[i]] then
			return false
		end
	end
	return true
end

--[==[Returns a table containing <code>WikimediaLanguage</code> objects (see [[Module:wikimedia languages]]), which represent languages and their codes as they are used in Wikimedia projects for interwiki linking and such. More than one object may be returned, as a single Wiktionary language may correspond to multiple Wikimedia languages. For example, Wiktionary's single code <code>sh</code> (Serbo-Croatian) maps to four Wikimedia codes: <code>sh</code> (Serbo-Croatian), <code>bs</code> (Bosnian), <code>hr</code> (Croatian) and <code>sr</code> (Serbian).
The code for the Wikimedia language is retrieved from the <code>wikimedia_codes</code> property in the data modules. If that property is not present, the code of the current language is used. If none of the available codes is actually a valid Wikimedia code, an empty table is returned.]==]
function Language:getWikimediaLanguages()
	local wm_langs = self._wikimediaLanguageObjects
	if wm_langs == nil then
		local get_wm_lang = require("Module:wikimedia languages").getByCode
		local codes = self:getWikimediaLanguageCodes()
		wm_langs = {}
		for i = 1, #codes do
			wm_langs[i] = get_wm_lang(codes[i])
		end
		self._wikimediaLanguageObjects = wm_langs
	end
	return wm_langs
end

function Language:getWikimediaLanguageCodes()
	local wm_langs = self._wikimediaLanguageCodes
	if wm_langs == nil then
		wm_langs = self._rawData.wikimedia_codes
		wm_langs = wm_langs and split(wm_langs, ",", true, true) or {self._code}
		self._wikimediaLanguageCodes = wm_langs
	end
	return wm_langs
end

--[==[
Returns the name of the Wikipedia article for the language. `project` specifies the language and project to retrieve
the article from, defaulting to {"enwiki"} for the English Wikipedia. Normally if specified it should be the project
code for a specific-language Wikipedia e.g. "zhwiki" for the Chinese Wikipedia, but it can be any project, including
non-Wikipedia ones. If the project is the English Wikipedia and the property {wikipedia_article} is present in the data
module it will be used first. In all other cases, a sitelink will be generated from {:getWikidataItem} (if set). The
resulting value (or lack of value) is cached so that subsequent calls are fast. If no value could be determined, and
`noCategoryFallback` is {false}, {:getCategoryName} is used as fallback; otherwise, {nil} is returned. Note that if
`noCategoryFallback` is {nil} or omitted, it defaults to {false} if the project is the English Wikipedia, otherwise
to {true}. In other words, under normal circumstances, if the English Wikipedia article couldn't be retrieved, the
return value will fall back to a link to the language's category, but this won't normally happen for any other project.
]==]
function Language:getWikipediaArticle(noCategoryFallback, project)
	project = project or "enwiki"
	local cached_value
	if project == "enwiki" then
		cached_value = self._wikipedia_article
		if cached_value == nil then
			cached_value = self._rawData.wikipedia_article
		end
	else
		-- If the project isn't enwiki, default to no category fallback, but this can be overridden by specifying the
		-- value `false`.
		if noCategoryFallback == nil then
			noCategoryFallback = true
		end
		if self._non_en_wikipedia_articles == nil then
			self._non_en_wikipedia_articles = {}
		end
		cached_value = self._non_en_wikipedia_articles[project]
	end
	if cached_value == nil then -- not false
		if self:getWikidataItem() and mw.wikibase then
			cached_value = mw.wikibase.sitelink(self:getWikidataItem(), project)
		end
		if not cached_value then
			cached_value = false
		end
	end
	-- Now cache the determined value.
	if project == "enwiki" then
		self._wikipedia_article = cached_value
	else
		self._non_en_wikipedia_articles[project] = cached_value
	end
	if not cached_value and not noCategoryFallback then
		return self:getCategoryName():gsub("Creole language", "Creole")
	end
	return cached_value or nil
end

function Language:makeWikipediaLink()
	return "[[w:" .. self:getWikipediaArticle() .. "|" .. self:getCanonicalName() .. "]]"
end

--[==[Returns the Wikidata item id for the language or <code>nil</code>. This corresponds to the the second field in the data modules.]==]
function Language:getWikidataItem()
	local item = self._WikidataItem
	if item == nil then
		item = self._rawData[2]
		-- If the value is nil, it's cached as false.
		item = item ~= nil and (type(item) == "number" and "Q" .. item or item) or false
		self._WikidataItem = item
	end
	return item or nil
end

--[==[Returns a table of <code>Script</code> objects for all scripts that the language is written in. See [[Module:scripts]].]==]
function Language:getScripts()
	local scripts = self._scriptObjects
	if scripts == nil then
		local codes = self:getScriptCodes()
		if codes[1] == "All" then
			scripts = self:loadData("Module:scripts/data")
		else
			local get_script = require("Module:scripts").getByCode
			scripts = {}
			for i = 1, #codes do
				scripts[i] = get_script(codes[i], nil, nil, self._useRequire)
			end
		end
		self._scriptObjects = scripts
	end
	return scripts
end

--[==[Returns the table of script codes in the language's data file.]==]
function Language:getScriptCodes()
	local scripts = self._scriptCodes
	if scripts == nil then
		scripts = self._rawData[4]
		if scripts then
			local codes, n = {}, 0
			for code in gmatch(scripts, "[^,]+") do
				n = n + 1
				-- Special handling of "Hants", which represents "Hani", "Hant" and "Hans" collectively.
				if code == "Hants" then
					codes[n] = "Hani"
					codes[n + 1] = "Hant"
					codes[n + 2] = "Hans"
					n = n + 2
				else
					codes[n] = code
				end
			end
			scripts = codes
		else
			scripts = {"None"}
		end
		self._scriptCodes = scripts
	end
	return scripts
end

--[==[Given some text, this function iterates through the scripts of a given language and tries to find the script that best matches the text. It returns a {{code|lua|Script}} object representing the script. If no match is found at all, it returns the {{code|lua|None}} script object.]==]
function Language:findBestScript(text, forceDetect)
	local useRequire = self._useRequire
	
	if not text or text == "" or text == "-" then
		return require("Module:scripts").getByCode("None", nil, nil, useRequire)
	end
	
	-- Differs from table returned by getScriptCodes, as Hants is not normalized into its constituents.
	local codes = self._bestScriptCodes
	if codes == nil then
		codes = self._rawData[4]
		codes = codes and split(codes, ",", true, true) or {"None"}
		self._bestScriptCodes = codes
	end
	
	local first_sc = codes[1]
	
	if first_sc == "All" then
		return require("Module:scripts").findBestScriptWithoutLang(text)
	end
	
	local get_script = require("Module:scripts").getByCode
	local codes_len = #codes
	
	if not (forceDetect or first_sc == "Hants" or codes_len > 1) then
		first_sc = get_script(first_sc, nil, nil, useRequire)
		local charset = first_sc.characters
		return charset and umatch(text, "[" .. charset .. "]") and first_sc or
			get_script("None", nil, nil, useRequire)
	end
	
	-- Remove all formatting characters.
	text = require("Module:utilities").get_plaintext(text)
	
	-- Remove all spaces and any ASCII punctuation. Some non-ASCII punctuation is script-specific, so can't be removed.
	text = ugsub(text, "[%s!\"#%%&'()*,%-./:;?@[\\%]_{}]+", "")
	if #text == 0 then
		return get_script("None", nil, nil, useRequire)
	end
	
	-- Try to match every script against the text,
	-- and return the one with the most matching characters.
	local bestcount, bestscript, length = 0
	for i = 1, codes_len do
		local sc = codes[i]
		-- Special case for "Hants", which is a special code that represents whichever of "Hant" or "Hans" best matches, or "Hani" if they match equally. This avoids having to list all three. In addition, "Hants" will be treated as the best match if there is at least one matching character, under the assumption that a Han script is desirable in terms that contain a mix of Han and other scripts (not counting those which use Jpan or Kore).
		if sc == "Hants" then
			local Hani = get_script("Hani", nil, nil, useRequire)
			if not Hant_chars then
				Hant_chars = self:loadData("Module:zh/data/ts")
				Hans_chars = self:loadData("Module:zh/data/st")
			end
			local t, s, found = 0, 0
			-- This is faster than using mw.ustring.gmatch directly.
			for ch in gmatch(ugsub(text, "[" .. Hani.characters .. "]", "\255%0"), "\255(.[\128-\191]*)") do
				found = true
				if Hant_chars[ch] then
					t = t + 1
					if Hans_chars[ch] then
						s = s + 1
					end
				elseif Hans_chars[ch] then
					s = s + 1
				else
					t, s = t + 1, s + 1
				end
			end
			
			if found then
				if t == s then
					return Hani
				end
				return get_script(t > s and "Hant" or "Hans", nil, nil, useRequire)
			end
		else
			sc = get_script(sc, nil, nil, useRequire)
			
			if not length then
				length = ulen(text)
			end
			
			-- Count characters by removing everything in the script's charset and comparing to the original length.
			local charset = sc.characters
			local count = charset and length - ulen(ugsub(text, "[" .. charset .. "]+", "")) or 0
			
			if count >= length then
				return sc
			elseif count > bestcount then
				bestcount = count
				bestscript = sc
			end
		end
	end
	
	-- Return best matching script, or otherwise None.
	return bestscript or get_script("None", nil, nil, useRequire)
end

--[==[Returns a <code>Family</code> object for the language family that the language belongs to. See [[Module:families]].]==]
function Language:getFamily()
	local family = self._familyObject
	if family == nil then
		family = self:getFamilyCode()
		-- If the value is nil, it's cached as false.
		family = family and require("Module:families").getByCode(family, self._useRequire) or false
		self._familyObject = family
	end
	return family or nil
end

--[==[Returns the family code in the language's data file.]==]
function Language:getFamilyCode()
	local family = self._familyCode
	if family == nil then
		-- If the value is nil, it's cached as false.
		family = self._rawData[3] or false
		self._familyCode = family
	end
	return family or nil
end

function Language:getFamilyName()
	local family = self._familyName
	if family == nil then
		family = self:getFamily()
		-- If the value is nil, it's cached as false.
		family = family and family:getCanonicalName() or false
		self._familyName = family
	end
	return family or nil
end

--[==[Check whether the language belongs to `family` (which can be a family code or object). A list of objects can be given in place of `family`; in that case, return true if the language belongs to any of the specified families. Note that some languages (in particular, certain creoles) can have multiple immediate ancestors potentially belonging to different families; in that case, return true if the language belongs to any of the specified families.]==]
function Language:inFamily(...)
	--check_object("family", nil, ...)
	for _, family in ipairs{...} do
		if type(family) == "table" then
			family = family:getCode()
		end
		local self_family_code = self:getFamilyCode()
		if not self_family_code then
			return false
		elseif self_family_code == family then
			return true
		end
		local self_family = self:getFamily()
		if self_family:inFamily(family) then
			return true
		-- If the family isn't a real family (e.g. creoles) check any ancestors.
		elseif self_family:getFamilyCode() == "qfa-not" then
			local ancestors = self:getAncestors()
			for _, ancestor in ipairs(ancestors) do
				if ancestor:inFamily(family) then
					return true
				end
			end
		end
	end
	return false
end

function Language:getParent()
	local parent = self._parentObject
	if parent == nil then
		parent = self:getParentCode()
		-- If the value is nil, it's cached as false.
		parent = parent and export.getByCode(parent, nil, true, true, self._useRequire) or false
		self._parentObject = parent
	end
	return parent or nil
end

function Language:getParentCode()
	local parent = self._parentCode
	if parent == nil then
		-- If the value is nil, it's cached as false.
		parent = self._rawData[5] or false
		self._parentCode = parent
	end
	return parent or nil
end

function Language:getParentName()
	local parent = self._parentName
	if parent == nil then
		parent = self:getParent()
		-- If the value is nil, it's cached as false.
		parent = parent and parent:getCanonicalName() or false
		self._parentName = parent
	end
	return parent or nil
end

function Language:getParentChain()
	local chain = self._parentChain
	if chain == nil then
		chain = {}
		local parent, n = self:getParent(), 0
		while parent do
			n = n + 1
			chain[n] = parent
			parent = parent:getParent()
		end
		self._parentChain = chain
	end
	return chain
end

function Language:hasParent(...)
	--check_object("language", nil, ...)
	for _, otherlang in ipairs{...} do
		for _, parent in ipairs(self:getParentChain()) do
			if type(otherlang) == "string" then
				if otherlang == parent:getCode() then return true end
			else
				if otherlang:getCode() == parent:getCode() then return true end
			end
		end
	end
	return false
end

--[==[
If the language is etymology-only, this iterates through parents until a full language or family is found, and the
corresponding object is returned. If the language is a full language, then it simply returns itself.
]==]
function Language:getFull()
	local full = self._fullObject
	if full == nil then
		full = self:getFullCode()
		full = full == self._code and self or
			export.getByCode(full, nil, nil, nil, self._useRequire)
		self._fullObject = full
	end
	return full
end

--[==[
If the language is an etymology-only language, this iterates through parents until a full language or family is
found, and the corresponding code is returned. If the language is a full language, then it simply returns the
language code.
]==]
function Language:getFullCode()
	return self._fullCode or self._code
end

--[==[
If the language is an etymology-only language, this iterates through parents until a full language or family is
found, and the corresponding canonical name is returned. If the language is a full language, then it simply returns
the canonical name of the language.
]==]
function Language:getFullName()
	local full = self._fullName
	if full == nil then
		full = self:getFull():getCanonicalName()
		self._fullName = full
	end
	return full
end

--[==[Returns a table of <code class="nf">Language</code> objects for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.]==]
function Language:getAncestors()
	if not self._ancestorObjects then
		self._ancestorObjects = {}
		local ancestors = shallowcopy(self:getAncestorCodes())
		if #ancestors > 0 then
			for _, ancestor in ipairs(ancestors) do
				insert(self._ancestorObjects, export.getByCode(ancestor, nil, true, nil, self._useRequire))
			end
		else
			local fam = self:getFamily()
			local protoLang = fam and fam:getProtoLanguage() or nil
			-- For the cases where the current language is the proto-language
			-- of its family, or an etymology-only language that is ancestral to that
			-- proto-language, we need to step up a level higher right from the
			-- start.
			if protoLang and (
				protoLang:getCode() == self._code or
				(self:hasType("etymology-only") and protoLang:hasAncestor(self))
			) then
				fam = fam:getFamily()
				protoLang = fam and fam:getProtoLanguage() or nil
			end
			while not protoLang and not (not fam or fam:getCode() == "qfa-not") do
				fam = fam:getFamily()
				protoLang = fam and fam:getProtoLanguage() or nil
			end
			insert(self._ancestorObjects, protoLang)
		end
	end
	return self._ancestorObjects
end

do
	-- Avoid a language being its own ancestor via class inheritance. We only need to check for this if the language has inherited an ancestor table from its parent, because we never want to drop ancestors that have been explicitly set in the data.
	-- Recursively iterate over ancestors until we either find self or run out. If self is found, return true.
	local function check_ancestor(self, lang)
		local codes = lang:getAncestorCodes()
		if not codes then
			return nil
		end
		for i = 1, #codes do
			local code = codes[i]
			if code == self._code then
				return true
			end
			local anc = export.getByCode(code, nil, true, nil, self._useRequire)
			if check_ancestor(self, anc) then
				return true
			end
		end
	end

	--[==[Returns a table of <code class="nf">Language</code> codes for all languages that this language is directly descended from. Generally this is only a single language, but creoles, pidgins and mixed languages can have multiple ancestors.]==]
	function Language:getAncestorCodes()
		if self._ancestorCodes then
			return self._ancestorCodes
		end
		local codes = self._rawData.ancestors
		if not codes then
			codes = {}
			self._ancestorCodes = codes
			return codes
		end
		codes = split(codes, ",", true, true)
		self._ancestorCodes = codes
		if (
			#codes == 0 or
			#self._stack == 1 or
			self._stack[#self._stack].ancestors
		) then
			return codes
		end
		local i, code = 1
		while i <= #codes do
			code = codes[i]
			if check_ancestor(self, self) then
				remove(codes, i)
			else
				i = i + 1
			end
		end
		return codes
	end
end

--[==[Given a list of language objects or codes, returns true if at least one of them is an ancestor. This includes any etymology-only children of that ancestor. If the language's ancestor(s) are etymology-only languages, it will also return true for those language parent(s) (e.g. if Vulgar Latin is the ancestor, it will also return true for its parent, Latin). However, a parent is excluded from this if the ancestor is also ancestral to that parent (e.g. if Classical Persian is the ancestor, Persian would return false, because Classical Persian is also ancestral to Persian).]==]
function Language:hasAncestor(...)
	--check_object("language", nil, ...)

	local function iterateOverAncestorTree(node, func, parent_check)
		local ancestors = node:getAncestors()
		local ancestorsParents = {}
		for _, ancestor in ipairs(ancestors) do
			local ret = func(ancestor) or iterateOverAncestorTree(ancestor, func, parent_check)
			if ret then return ret end
		end
		-- Check the parents of any ancestors. We don't do this if checking the parents of the other language, so that we exclude any etymology-only children of those parents that are not directly related (e.g. if the ancestor is Vulgar Latin and we are checking New Latin, we want it to return false because they are on different ancestral branches. As such, if we're already checking the parent of New Latin (Latin) we don't want to compare it to the parent of the ancestor (Latin), as this would be a false positive; it should be one or the other).
		if not parent_check then
			return nil
		end
		for _, ancestor in ipairs(ancestors) do
			local ancestorParents = ancestor:getParentChain()
			for _, ancestorParent in ipairs(ancestorParents) do
				if ancestorParent:getCode() == self._code or ancestorParent:hasAncestor(ancestor) then
					break
				else
					insert(ancestorsParents, ancestorParent)
				end
			end
		end
		for _, ancestorParent in ipairs(ancestorsParents) do
			local ret = func(ancestorParent)
			if ret then return ret end
		end
	end

	local function do_iteration(otherlang, parent_check)
		-- otherlang can't be self
		if (type(otherlang) == "string" and otherlang or otherlang:getCode()) == self._code then
			return false
		end
		repeat
			if iterateOverAncestorTree(
				self,
				function(ancestor)
					return ancestor:getCode() == (type(otherlang) == "string" and otherlang or otherlang:getCode())
				end,
				parent_check
			) then
				return true
			elseif type(otherlang) == "string" then
				otherlang = export.getByCode(otherlang, nil, true, nil, self._useRequire)
			end
			otherlang = otherlang:getParent()
			parent_check = false
		until not otherlang
	end

	local parent_check = true
	for _, otherlang in ipairs{...} do
		local ret = do_iteration(otherlang, parent_check)
		if ret then
			return true
		end
	end
	return false
end

function Language:getAncestorChain()
	if not self._ancestorChain then
		self._ancestorChain = {}
		local step = self
		while true do
			local ancestors = step:getAncestors()
			step = #ancestors == 1 and ancestors[1] or nil
			if not step then break end
			insert(self._ancestorChain, 1, step)
		end
	end
	return self._ancestorChain
end

local function fetch_descendants(self, format)
	local languages = require("Module:languages/code to canonical name")
	local etymology_languages = require("Module:etymology languages/code to canonical name")
	local families = require("Module:families/code to canonical name")
	local descendants = {}
	local family = self:getFamily()
	-- Iterate over all three datasets.
	for _, data in ipairs{languages, etymology_languages, families} do
		for code in pairs(data) do
			local lang = export.getByCode(code, nil, true, true, self._useRequire)
			-- Test for a descendant. Earlier tests weed out most candidates, while the more intensive tests are only used sparingly.
			if (
				code ~= self._code and -- Not self.
				lang:inFamily(family) and -- In the same family.
				(
					family:getProtoLanguageCode() == self._code or -- Self is the protolanguage.
					self:hasDescendant(lang) or -- Full hasDescendant check.
					(lang:getFullCode() == self._code and not self:hasAncestor(lang)) -- Etymology-only child which isn't an ancestor.
				)
			) then
				if format == "object" then
					insert(descendants, lang)
				elseif format == "code" then
					insert(descendants, code)
				elseif format == "name" then
					insert(descendants, lang:getCanonicalName())
				end
			end
		end
	end
	return descendants
end

function Language:getDescendants()
	if not self._descendantObjects then
		self._descendantObjects = fetch_descendants(self, "object")
	end
	return self._descendantObjects
end

function Language:getDescendantCodes()
	if not self._descendantCodes then
		self._descendantCodes = fetch_descendants(self, "code")
	end
	return self._descendantCodes
end

function Language:getDescendantNames()
	if not self._descendantNames then
		self._descendantNames = fetch_descendants(self, "name")
	end
	return self._descendantNames
end

function Language:hasDescendant(...)
	for _, lang in ipairs{...} do
		if type(lang) == "string" then
			lang = export.getByCode(lang, nil, true, nil, self._useRequire)
		end
		if lang:hasAncestor(self) then
			return true
		end
	end
	return false
end

local function fetch_children(self, format)
	local m_etym_data = require("Module:etymology languages/data")
	local self_code = self._code
	local children = {}
	for code, data in pairs(m_etym_data) do
		local _data = data
		repeat
			local parent = _data[5]
			if parent == self_code then
				if format == "object" then
					insert(children, export.getByCode(code, nil, true, nil, self._useRequire))
				elseif format == "code" then
					insert(children, code)
				elseif format == "name" then
					insert(children, data[1])
				end
				break
			end
			_data = m_etym_data[parent]
		until not _data
	end
	return children
end

function Language:getChildren()
	if not self._childObjects then
		self._childObjects = fetch_children(self, "object")
	end
	return self._childObjects
end

function Language:getChildrenCodes()
	if not self._childCodes then
		self._childCodes = fetch_children(self, "code")
	end
	return self._childCodes
end

function Language:getChildrenNames()
	if not self._childNames then
		self._childNames = fetch_children(self, "name")
	end
	return self._childNames
end

function Language:hasChild(...)
	local lang = ...
	if not lang then
		return false
	elseif type(lang) == "string" then
		lang = export.getByCode(lang, nil, true, nil, self._useRequire)
	end
	if lang:hasParent(self) then
		return true
	end
	return self:hasChild(select(2, ...))
end

--[==[Returns the name of the main category of that language. Example: {{code|lua|"French language"}} for French, whose category is at [[:Category:French language]]. Unless optional argument <code>nocap</code> is given, the language name at the beginning of the returned value will be capitalized. This capitalization is correct for category names, but not if the language name is lowercase and the returned value of this function is used in the middle of a sentence.]==]
function Language:getCategoryName(nocap)
	if not self._categoryName then
		local name = self:getCanonicalName()
		-- Only add " language" if a full language.
		if #self._stack == 1 then
			-- If the name already has "language" in it, don't add it.
			if not name:match("[Ll]anguage$") then
				name = name .. " language"
			end
		end
		self._categoryName = name
	end
	if nocap then
		return self._categoryName
	else
		return mw.getContentLanguage():ucfirst(self._categoryName)
	end
end

--[==[Creates a link to the category; the link text is the canonical name.]==]
function Language:makeCategoryLink()
	return "[[:Category:" .. self:getCategoryName() .. "|" .. self:getDisplayForm() .. "]]"
end

function Language:getStandardCharacters(sc)
	if type(self._rawData.standardChars) ~= "table" then
		return self._rawData.standardChars
	else
		if sc and type(sc) ~= "string" then
			check_object("script", nil, sc)
			sc = sc:getCode()
		end
		if (not sc) or sc == "None" then
			local scripts = {}
			for _, script in pairs(self._rawData.standardChars) do
				insert(scripts, script)
			end
			return concat(scripts)
		end
		if self._rawData.standardChars[sc] then
			return self._rawData.standardChars[sc] .. (self._rawData.standardChars[1] or "")
		end
	end
end

--[==[Make the entry name (i.e. the correct page name).]==]
function Language:makeEntryName(text, sc)
	if (not text) or text == "" then
		return text, nil, {}
	end
	
	-- Set `unsupported` as true if certain conditions are met.
	local unsupported
	-- Check if there's an unsupported character. \239\191\189 is the replacement character U+FFFD, which can't be typed directly here due to an abuse filter. Unix-style dot-slash notation is also unsupported, as it is used for relative paths in links, as are 3 or more consecutive tildes.
	if (
		text:find("[#<>%[%]_{|}]") or
		text:find("\239\191\189") or
		text:find("%f[^%z/]%.%.?%f[%z/]") or
		text:find("~~~")
	) then
		unsupported = true
	-- If it looks like an interwiki link.
	elseif text:find(":") then
		local prefix = text:gsub("^:*(.-):.*", string.ulower)
		if (
			self:loadData("Module:data/namespaces")[prefix] or
			self:loadData("Module:data/interwikis")[prefix]
		) then
			unsupported = true
		end
	end

	-- Check if the text is a listed unsupported title.
	local unsupportedTitles = self:loadData("Module:links/data").unsupported_titles
	if unsupportedTitles[text] then
		return "Unsupported titles/" .. unsupportedTitles[text], nil, {}
	end

	sc = checkScript(text, self, sc)

	local fail, cats
	text = normalize(text, sc)
	text, fail, cats = iterateSectionSubstitutions(text, nil, nil, self, sc, self._rawData.entry_name, "makeEntryName")

	text = umatch(text, "^[¿¡]?(.-[^%s%p].-)%s*[؟?!;՛՜ ՞ ՟？！︖︕।॥။၊་།]?$") or text


	-- Escape unsupported characters so they can be used in titles. ` is used as a delimiter for this, so a raw use of it in an unsupported title is also escaped here to prevent interference; this is only done with unsupported titles, though, so inclusion won't in itself mean a title is treated as unsupported (which is why it's excluded from the earlier test).
	if unsupported then
		local unsupported_characters = self:loadData("Module:links/data").unsupported_characters
		text = text
			:gsub("[#<>%[%]_`{|}\239]\191?\189?", unsupported_characters)
			:gsub("%f[^%z/]%.%.?%f[%z/]", function(m)
				return m:gsub("%.", "`period`")
			end)
			:gsub("~~~+", function(m)
				return m:gsub("~", "`tilde`")
			end)
		text = "Unsupported titles/" .. text
	end

	return text, fail, cats
end

--[==[Generates alternative forms using a specified method, and returns them as a table. If no method is specified, returns a table containing only the input term.]==]
function Language:generateForms(text, sc)
	if self._rawData.generate_forms then
		sc = checkScript(text, self, sc)
		return require("Module:" .. self._rawData.generate_forms).generateForms(text, self._code, sc:getCode())
	else
		return {text}
	end
end

--[==[Creates a sort key for the given entry name, following the rules appropriate for the language. This removes diacritical marks from the entry name if they are not considered significant for sorting, and may perform some other changes. Any initial hyphen is also removed, and anything parentheses is removed as well.
The <code>sort_key</code> setting for each language in the data modules defines the replacements made by this function, or it gives the name of the module that takes the entry name and returns a sortkey.]==]
function Language:makeSortKey(text, sc)
	if (not text) or text == "" then
		return text, nil, {}
	end
	if text:find("<[^<>]+>") then
		track("track HTML tag")
	end
	-- Remove directional characters, soft hyphens, strip markers and HTML tags.
	text = ugsub(text, "[\194\173\226\128\170-\226\128\174\226\129\166-\226\129\169]", "")
	text = mw.text.unstrip(text)
		:gsub("<[^<>]+>", "")

	text = decode_uri(text, "PATH")
	text = checkNoEntities(self, text)

	-- Remove initial hyphens and * unless the term only consists of spacing + punctuation characters.
	text = ugsub(text, "^([􀀀-􏿽]*)[-־ـ᠊*]+([􀀀-􏿽]*)(.*[^%s%p].*)", "%1%2%3")

	sc = checkScript(text, self, sc)

	text = normalize(text, sc)
	text = removeCarets(text, sc)

	-- For languages with dotted dotless i, ensure that "İ" is sorted as "i", and "I" is sorted as "ı".
	if self:hasDottedDotlessI() then
		text = text:gsub("I\204\135", "i") -- decomposed "İ"
			:gsub("I", "ı")
		text = sc:toFixedNFD(text)
	end
	-- Convert to lowercase, make the sortkey, then convert to uppercase. Where the language has dotted dotless i, it is usually not necessary to convert "i" to "İ" and "ı" to "I" first, because "I" will always be interpreted as conventional "I" (not dotless "İ") by any sorting algorithms, which will have been taken into account by the sortkey substitutions themselves. However, if no sortkey substitutions have been specified, then conversion is necessary so as to prevent "i" and "ı" both being sorted as "I".
	-- An exception is made for scripts that (sometimes) sort by scraping page content, as that means they are sensitive to changes in capitalization (as it changes the target page).
	local fail, cats
	if not sc:sortByScraping() then
		text = text:ulower()
	end

	text, fail, cats = iterateSectionSubstitutions(text, nil, nil, self, sc, self._rawData.sort_key, "makeSortKey")

	if not sc:sortByScraping() then
		if self:hasDottedDotlessI() and not self._rawData.sort_key then
			text = text:gsub("ı", "I")
				:gsub("i", "İ")
			text = sc:toFixedNFC(text)
		end
		text = text:uupper()
	end

	-- Remove parentheses, as long as they are either preceded or followed by something.
	text = text
		:gsub("(.)[()]+", "%1")
		:gsub("[()]+(.)", "%1")

	text = escape_risky_characters(text)
	return text, fail, cats
end

--[==[Create the form used as as a basis for display text and transliteration.]==]
local function processDisplayText(text, self, sc, keepCarets, keepPrefixes)
	local subbedChars = {}
	text, subbedChars = doTempSubstitutions(text, subbedChars, keepCarets)

	text = decode_uri(text, "PATH")
	text = checkNoEntities(self, text)

	sc = checkScript(text, self, sc)
	local fail, cats
	text = normalize(text, sc)
	text, fail, cats, subbedChars = iterateSectionSubstitutions(text, subbedChars, keepCarets, self, sc, self._rawData.display_text, "makeDisplayText")

	text = removeCarets(text, sc)

	-- Remove any interwiki link prefixes (unless they have been escaped or this has been disabled).
	if text:match(":") and not keepPrefixes then
		local rep
		repeat
			text, rep = text:gsub("\\\\(\\*:)", "\3%1")
		until rep == 0
		text = text
			:gsub("\\:", "\4")
		while true do
			local prefix = text:gsub("^(.-):.+", function(m1)
				return m1:gsub("\244[\128-\191]*", "")
			end)
			if not prefix or prefix == text then
				break
			end
			local lower_prefix = prefix:ulower()
			if not (self:loadData("Module:data/interwikis")[lower_prefix] or prefix == "") then
				break
			end
			text = text:gsub("^(.-):(.*)", function(m1, m2)
				local ret = {}
				for subbedChar in gmatch(m1, "\244[\128-\191]*") do
					insert(ret, subbedChar)
				end
				return concat(ret) .. m2
			end)
		end
		text = text
			:gsub("\3", "\\")
			:gsub("\4", ":")
	end

	return text, fail, cats, subbedChars
end

--[==[Make the display text (i.e. what is displayed on the page).]==]
function Language:makeDisplayText(text, sc, keepPrefixes)
	if (not text) or text == "" then
		return text, nil, {}
	end
	
	local fail, cats, subbedChars
	text, fail, cats, subbedChars = processDisplayText(text, self, sc, nil, keepPrefixes)

	text = escape_risky_characters(text)
	return undoTempSubstitutions(text, subbedChars), fail, cats
end

--[==[Transliterates the text from the given script into the Latin script (see [[Wiktionary:Transliteration and romanization]]). The language must have the <code>translit</code> property for this to work; if it is not present, {{code|lua|nil}} is returned.
Returns three values:
# The transliteration.
# A boolean which indicates whether the transliteration failed for an unexpected reason. If {{code|lua|false}}, then the transliteration either succeeded, or the module is returning nothing in a controlled way (e.g. the input was {{code|lua|"-"}}). Generally, this means that no maintenance action is required. If {{code|lua|true}}, then the transliteration is {{code|lua|nil}} because either the input or output was defective in some way (e.g. [[Module:ar-translit]] will not transliterate non-vocalised inputs, and this module will fail partially-completed transliterations in all languages). Note that this value can be manually set by the transliteration module, so make sure to cross-check to ensure it is accurate.
# A table of categories selected by the transliteration module, which should be in the format expected by {{code|lua|format_categories}} in [[Module:utilities]].
The <code>sc</code> parameter is handled by the transliteration module, and how it is handled is specific to that module. Some transliteration modules may tolerate {{code|lua|nil}} as the script, others require it to be one of the possible scripts that the module can transliterate, and will show an error if it's not one of them. For this reason, the <code>sc</code> parameter should always be provided when writing non-language-specific code.
The <code>module_override</code> parameter is used to override the default module that is used to provide the transliteration. This is useful in cases where you need to demonstrate a particular module in use, but there is no default module yet, or you want to demonstrate an alternative version of a transliteration module before making it official. It should not be used in real modules or templates, only for testing. All uses of this parameter are tracked by [[Wiktionary:Tracking/module_override]].
'''Known bugs''':
* This function assumes {tr(s1) .. tr(s2) == tr(s1 .. s2)}. When this assertion fails, wikitext markups like <nowiki>'''</nowiki> can cause wrong transliterations.
* HTML entities like <code>&amp;apos;</code>, often used to escape wikitext markups, do not work.]==]
function Language:transliterate(text, sc, module_override)
	-- If there is no text, or the language doesn't have transliteration data and there's no override, return nil.
	if not (self._rawData.translit or module_override) then
		return nil, false, {}
	elseif (not text) or text == "" or text == "-" then
		return text, false, {}
	end
	-- If the script is not transliteratable (and no override is given), return nil.
	sc = checkScript(text, self, sc)
	if not (sc:isTransliterated() or module_override) then
		return nil, true, {}
	end

	-- Remove any strip markers.
	text = mw.text.unstrip(text)

	-- Get the display text with the keepCarets flag set.
	local fail, cats, subbedChars
	text, fail, cats, subbedChars = processDisplayText(text, self, sc, true)

	-- Transliterate (using the module override if applicable).
	text, fail, cats, subbedChars = iterateSectionSubstitutions(text, subbedChars, true, self, sc, module_override or self._rawData.translit, "tr")
	
	if not text then
		return nil, true, cats
	end
	
	-- Incomplete transliterations return nil.
	local charset = sc.characters
	if charset and umatch(text, "[" .. charset .. "]") then
		-- Remove any characters in Latin, which includes Latin characters also included in other scripts (as these are false positives). Anything remaining should only be script code "None" (e.g. numerals).
		local check_text = ugsub(text, "[" .. require("Module:scripts").getByCode("Latn").characters .. "]", "")
		if require("Module:scripts").findBestScriptWithoutLang(check_text):getCode() ~= "None" then
			return nil, true, cats
		end
	end

	text = escape_risky_characters(text)
	text = undoTempSubstitutions(text, subbedChars)

	-- If the script does not use capitalization, then capitalize any letters of the transliteration which are immediately preceded by a caret (and remove the caret).
	if text and not sc:hasCapitalization() and text:match("%^") then
		text = processCarets(text, "%^([\128-\191\244]*%*?)([^\128-\191\244][\128-\191]*)", function(m1, m2)
			return m1 .. m2:uupper()
		end)
	end

	-- Track module overrides.
	if module_override ~= nil then
		track("module_override")
	end

	fail = text == nil and (not not fail) or false

	return text, fail, cats
end

function Language:overrideManualTranslit()
	return not not self._rawData.override_translit
end

--[==[Returns {{code|lua|true}} if the language has a transliteration module, or {{code|lua|false}} if it doesn't.]==]
function Language:hasTranslit()
	return not not self._rawData.translit
end

function Language:link_tr()
	return not not self._rawData.link_tr
end

--[==[Returns {{code|lua|true}} if the language uses the letters I/ı and İ/i, or {{code|lua|false}} if it doesn't.]==]
function Language:hasDottedDotlessI()
	return not not self._rawData.dotted_dotless_i
end

function Language:toJSON(returnTable)
	local entryNamePatterns = nil
	local entryNameRemoveDiacritics = nil

	if self._rawData.entry_name then
		entryNameRemoveDiacritics = self._rawData.entry_name.remove_diacritics
		if self._rawData.entry_name.from then
			entryNamePatterns = {}
			for i, from in ipairs(self._rawData.entry_name.from) do
				insert(entryNamePatterns, {from = from, to = self._rawData.entry_name.to[i] or ""})
			end
		end
	end
	
	-- mainCode should only end up non-nil if dontCanonicalizeAliases is passed to make_object().
	local ret = m_table.deepcopy{
		ancestors = self:getAncestorCodes(),
		canonicalName = self:getCanonicalName(),
		categoryName = self:getCategoryName("nocap"),
		code = self._code,
		mainCode = self._main_code,
		entryNamePatterns = entryNamePatterns,
		entryNameRemoveDiacritics = entryNameRemoveDiacritics,
		family = self:getFamilyCode(),
		otherNames = self:getOtherNames(true),
		aliases = self:getAliases(),
		varieties = self:getVarieties(),
		scripts = self:getScriptCodes(),
		parent = self._parentCode or nil,
		full = self._fullCode or nil,
		type = m_table.keysToList(self:getTypes()),
		wikimediaLanguages = self:getWikimediaLanguageCodes(),
		wikidataItem = self:getWikidataItem(),
	}

	if returnTable then
		return ret
	else
		return require("Module:JSON").toJSON(ret)
	end
end

--[==[
<span style="color: #BA0000">This function is not for use in entries or other content pages.</span>
Returns a blob of data about the language. The format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.
-- Do NOT use these methods!
-- All uses should be pre-approved on the talk page!
]==]
function Language:getRawData()
	local rawData = {}
	for _, element in ipairs(self._stack) do
		for k, v in pairs(element) do
			rawData[k] = v
		end
	end
	return rawData
end

--[==[<span style="color: #BA0000">This function is not for use in entries or other content pages.</span>
Returns a blob of data about the language that contains the "extra data". Much like with getRawData, the format of this blob is undocumented, and perhaps unstable; it's intended for things like the module's own unit-tests, which are "close friends" with the module and will be kept up-to-date as the format changes.]==]
function Language:getRawExtraData()
	if #self._stack == 1 then
		self:loadInExtraData()
	end
	return self._extraData
end

local function getRawExtraLanguageData(self, code)
	local modulename = export.getExtraDataModuleName(code)
	return modulename and self:loadData("Module:" .. modulename)[code] or nil
end

function Language:loadInExtraData()
	if not self._extraData then
		-- load extra data from module and assign to _extraData field
		-- use empty table as a fallback if extra data is nil
		self._extraData = getRawExtraLanguageData(self, self._code) or {}
	end
end

function export.getDataModuleName(code)
	if code:match("^%l%l$") then
		return "languages/data/2"
	elseif code:match("^%l%l%l$") then
		local prefix = code:sub(1, 1)
		return "languages/data/3/" .. prefix
	elseif code:match("^[%l-]+$") then
		return "languages/data/exceptional"
	else
		return nil
	end
end

function export.getExtraDataModuleName(code)
	local dataModule = export.getDataModuleName(code)
	return dataModule and dataModule .. "/extra" or nil
end

do
	local key_types = {
		[2] = "unique",
		aliases = "unique",
		otherNames = "unique",
		type = "append",
		varieties = "unique"
	}

	local function make_stack(code, input_code, data, parent, useRequire)
		parent.__index = parent

		local lang = {
			_code = input_code,
			_useRequire = useRequire or nil
		}
		-- This can only happen if dontCanonicalizeAliases is passed to make_object().
		if code ~= input_code then
			lang._main_code = code
		end

		-- Full language.
		if not parent._stack then
			-- Create stack, accessed with rawData metamethod.
			local stack = parent._rawData and {parent._rawData, data} or {data}
			lang._stack = stack
			lang._rawData = setmetatable({}, {
				__index = function(t, k)
					local key_type = key_types[k]
					-- Data that isn't inherited from the parent.
					if key_type == "unique" then
						return stack[#stack][k]
					-- Data that is appended by each generation.
					elseif key_type == "append" then
						local parts = {}
						for i = 1, #stack do
							insert(parts, stack[i][k])
						end
						if type(parts[1]) == "string" then
							return concat(parts, ","), true
						end
					-- Otherwise, iterate down the stack, looking for a match.
					else
						local i = #stack
						while not stack[i][k] and i > 1 do
							i = i - 1
						end
						return stack[i][k]
					end
				end,
				-- Retain immutability (as writing to rawData will break functionality).
				__newindex = function()
					error("not allowed to edit rawData")
				end
			})
			-- Full code is the parent code.
			lang._fullCode = parent._code or code
		-- Etymology-only.
		else
			-- Copy over rawData and stack to the new object, and add new layer to stack.
			lang._rawData = parent._rawData
			lang._stack = parent._stack
			insert(lang._stack, data)
			-- Copy full code.
			lang._fullCode = parent._fullCode
		end

		return setmetatable(lang, parent)
	end

	function export.makeObject(code, data, useRequire, dontCanonicalizeAliases)
		if not data then
			return nil
		end

		-- Convert any aliases.
		local input_code = code
		code = normalize_code(code)
		input_code = dontCanonicalizeAliases and input_code or code

		if data.type:find("family") and not data[5] then
			return require("Module:families").makeObject(code, data, useRequire)
		else
			local parent
			if data[5] then
				parent = export.getByCode(data[5], nil, true, true, useRequire)
			else
				parent = Language
			end
			return make_stack(code, input_code, data, parent, useRequire)
		end
	end
end

--[==[Finds the language whose code matches the one provided. If it exists, it returns a <code class="nf">Language</code> object representing the language. Otherwise, it returns {{code|lua|nil}}, unless <code class="n">paramForError</code> is given, in which case an error is generated. If <code class="n">paramForError</code> is {{code|lua|true}}, a generic error message mentioning the bad code is generated; otherwise <code class="n">paramForError</code> should be a string or number specifying the parameter that the code came from, and this parameter will be mentioned in the error message along with the bad code. If <code class="n">allowEtymLang</code> is specified, etymology-only language codes are allowed and looked up along with normal language codes. If <code class="n">allowFamily</code> is specified, language family codes are allowed and looked up along with normal language codes.]==]
function export.getByCode(code, paramForError, allowEtymLang, allowFamily, useRequire)
	if type(code) ~= "string" then
		local typ
		if not code then
			typ = "nil"
		elseif check_object("language", true, code) then
			typ = "a language object"
		elseif check_object("family", true, code) then
			typ = "a family object"
		else
			typ = "a " .. type(code)
		end
		error("The function getByCode expects a string as its first argument, but received " .. typ .. ".")
	end

	local function conditionalRequire(modulename)
		if useRequire then
			return require(modulename)
		else
			return load_data(modulename)
		end
	end

	-- FIXME: Temporary. Lists bad codes to track, so we can consider eliminating them.
	-- We list them directly here rather than in a separate module (cf. [[Module:etymology languages/track-bad-etym-code]])
	-- in the hope that this reduces memory usage as we have to do this for every invocation of getByCode() for every
	-- language code.
	local codes_to_track = {
		-- Codes duplicated been full and etymology-only languages
		["bsg"] = true,
		["rdb"] = true,
		["tgf"] = true,
		-- Aliases actively being deprecated
		["prv"] = true, -- oc-pro
		-- Codes that will be converted to families
		["nan"] = true,
		-- Codes being renamed
		["cmn-wadegile"] = true,
		["wuu-ngb"] = true,
		["wuu-hzh"] = true,
		["wuu-szh"] = true,
	}

	local function track_bad_code(code)
		if codes_to_track[code] then
			track(code)
		end
		return true
	end

	local modulename = export.getDataModuleName(code)

	local function get_data(code)
		return modulename and
			track_bad_code(code) and conditionalRequire("Module:" .. modulename)[code] or
			(allowEtymLang and require("Module:etymology languages/track-bad-etym-code")(code) and conditionalRequire("Module:etymology languages/data")[code]) or
			(allowFamily and conditionalRequire("Module:families/data")[code]) or
			(allowEtymLang and allowFamily and require("Module:families/track-bad-etym-code")(code) and conditionalRequire("Module:families/data/etymology")[code])
	end

	local data = get_data(code) or get_data(normalize_code(code))

	local retval = code and data and export.makeObject(code, data, useRequire)

	if not retval and paramForError then
		require("Module:languages/errorGetBy").code(code, paramForError, allowEtymLang, allowFamily)
	end

	return retval
end

--[==[Finds the language whose canonical name (the name used to represent that language on Wiktionary) or other name matches the one provided. If it exists, it returns a <code class="nf">Language</code> object representing the language. Otherwise, it returns {{code|lua|nil}}, unless <code class="n">paramForError</code> is given, in which case an error is generated. If <code class="n">allowEtymLang</code> is specified, etymology-only language codes are allowed and looked up along with normal language codes. If <code class="n">allowFamily</code> is specified, language family codes are allowed and looked up along with normal language codes.
The canonical name of languages should always be unique (it is an error for two languages on Wiktionary to share the same canonical name), so this is guaranteed to give at most one result.
This function is powered by [[Module:languages/canonical names]], which contains a pre-generated mapping of full-language canonical names to codes. It is generated by going through the [[:Category:Language data modules]] for full languages. When <code class="n">allowEtymLang</code> is specified for the above function, [[Module:etymology languages/canonical names]] may also be used, and when <code class="n">allowFamily</code> is specified for the above function, [[Module:families/canonical names]] may also be used.]==]
function export.getByCanonicalName(name, errorIfInvalid, allowEtymLang, allowFamily, useRequire)
	local function conditionalRequire(modulename)
		if useRequire then
			return require(modulename)
		else
			return load_data(modulename)
		end
	end

	local byName = conditionalRequire("Module:languages/canonical names")
	local code = byName and byName[name]

	if not code and allowEtymLang then
		byName = conditionalRequire("Module:etymology languages/canonical names")
		code = byName and byName[name] or
			byName[name:gsub(" [Ss]ubstrate$", "")] or
			byName[name:gsub("^a ", "")] or
			byName[name:gsub("^a ", ""):gsub(" [Ss]ubstrate$", "")] or
			-- For etymology families like "ira-pro".
			-- FIXME: This is not ideal, as it allows " languages" to be appended to any etymology-only language, too.
			byName[name:match("^(.*) languages$")]
	end

	if not code and allowFamily then
		byName = conditionalRequire("Module:families/canonical names")
		code = byName and byName[name] or
			byName[name:match("^(.*) languages$")]
	end

	local retval = code and export.getByCode(code, errorIfInvalid, allowEtymLang, allowFamily, useRequire)

	if not retval and errorIfInvalid then
		require("Module:languages/errorGetBy").canonicalName(name, allowEtymLang, allowFamily)
	end

	return retval
end

--[==[Used by [[Module:languages/data/2]] (et al.) to add default types to the entities returned.]==]
function export.addDefaultTypes(data, regular, ...)
	local n = arg.n
	local types = n > 0 and concat(arg, ",") or ""
	for _, entity in next, data do
		-- "regular" encompasses everything that doesn't have another type already assigned.
		if regular then
			entity.type = entity.type or "regular"
		end
		if n > 0 then
			entity.type =  types .. (entity.type and ("," .. entity.type) or "")
		end
	end
	return data
end

--[==[Used by [[Module:languages/data/2]] (et al.) and [[Module:etymology languages/data]] to finalize language-related data into the format that is actually returned.]==]
function export.finalizeLanguageData(data)
	-- 4 is scripts.
	local fields = {4, "ancestors", "type", "wikimedia_codes"}
	local fields_len = #fields
	for _, entity in next, data do
		for i = 1, fields_len do
			local key = fields[i]
			local field = entity[key]
			if field then
				entity[key] = gsub(field, "%s+", "")
			end
		end
	end
	return data
end

--[==[Used by [[Module:etymology languages/data]] and [[Module:families/data/etymology]] to finalize etymology-related data into the format that is actually returned.]==]
function export.finalizeEtymologyData(data)
	local aliases = {}
	for _, entity in next, data do
		-- Move parent to 5 and family to 3.
		entity[5] = entity[3]
		entity[3] = entity.family
		entity.family = nil
	end
	for code, alias in next, aliases do
		data[code] = alias
	end
	return data
end

--[==[For backwards compatibility only; modules should require the error themselves.]==]
function export.err(lang_code, param, code_desc, template_tag, not_real_lang)
	return require("Module:languages/error")(lang_code, param, code_desc, template_tag, not_real_lang)
end

return export
Retrieved from "https://en.wiktionary.org/w/index.php?title=Module:languages&oldid=79061616"