Module:Linguistic

Revision as of 04:48, 26 October 2022 by wikimedia>Jonesey95 (require strict per TPER)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Documentation for this module may be created at Module:Linguistic/doc

--[[
  __  __           _       _        _     _                   _     _   _      
 |  \/  | ___   __| |_   _| | ___ _| |   (_)_ __   __ _ _   _(_)___| |_(_) ___ 
 | |\/| |/ _ \ / _` | | | | |/ _ (_) |   | | '_ \ / _` | | | | / __| __| |/ __|
 | |  | | (_) | (_| | |_| | |  __/_| |___| | | | | (_| | |_| | \__ \ |_| | (__ 
 |_|  |_|\___/ \__,_|\__,_|_|\___(_)_____|_|_| |_|\__, |\__,_|_|___/\__|_|\___|
                                                  |___/                        
Simple internationalization functions that can be called by other modules.
This Module was copied from Wikimedia Commons, so please request changes there.

Maintainers:
* Zolo - original version
* Jarekt

Dependencies:
- Module uses c:data:I18n/Or.tab
]]
require('strict')

-- ==================================================
-- === Internal functions ===========================
-- ==================================================

local function langSwitch(list,lang)
	local langList = mw.language.getFallbacksFor(lang)
	table.insert(langList,1,lang)
	for i,language in ipairs(langList) do
		if list[language] then
			return list[language]
		end
	end
	return nil
end

------------------------------------------------------------------------------
-- read Commons Data:SOMENAME.tab dataset and look for message identified by a 
-- "key" in a language "lang". See editAtWikidata as an example.
local function formatMessage(dataset, key, lang)
	for _, row in pairs(mw.ext.data.get(dataset, lang).data) do
		local id, msg = unpack(row)
		if id == key then
			return mw.message.newRawMessage(msg):plain()
		end
	end
	error('Invalid message key "' .. key .. '"')
end

------------------------------------------------------------------------------
local function nowiki(str) -- remove all the links
	if not str then
		return str
	end
	str = mw.ustring.gsub(str, "<[^>]*>", "")    -- remove all html tags from str
	str = mw.ustring.gsub(str, "'''", "")        -- remove bold 
	str = mw.ustring.gsub(str, "''", "")         -- remove italics
	str = mw.ustring.gsub(str, "%[%[[Ff]ile:[^%]]+%]%]", "") -- remove file icons
	str = mw.ustring.gsub(str, "%[%[[^|]*|", "") -- remove piped links, like "[[:en:test|"
	str = mw.ustring.gsub(str, "%[[^ ]+ ", "")   -- remove URL links, like "[https://www.wikidata.org/wiki/Q2706250 "
	str = mw.ustring.gsub(str, "%[%[", "" )      -- remove piped links, like "[["
	str = mw.ustring.gsub(str, "%]", "" )        -- remove piped links, like "]" or "]]"
	return str
end

------------------------------------------------------------------------------
-- Function allowing for consistent treatment of boolean-like wikitext input.
-- It works similarly to Module:Yesno
local function yesno(val, default)
	if type(val) == 'boolean' then
		return val
	elseif type(val) == 'number' then
		if val==1 then 
			return true
		elseif val==0 then
			return false
		end
	elseif type(val) == 'string' then
	    val = mw.ustring.lower(val)  -- put in lower case
	    if val == 'no'  or val == 'n' or val == 'false' or val == '0' then
	        return false
	    elseif val == 'yes' or val == 'y' or val == 'true'  or val == '1' then
	        return true
	    end
    end
    return default
end

-- ==================================================
-- === External functions ===========================
-- ==================================================
local p = {}

-- ===========================================================================
-- === Version of the function to be called from other LUA codes
-- ===========================================================================

------------------------------------------------------------------------------------------
function p.vowelfirst (str)
	if str then 
		local vowels = 'aeiouyąăẵằẳặȃắâẫấầẩậãäǟāáàȁǎảẚåǻḁạǡæǣǽĕȇêễếềểệḙẽḛëēḕéḗèȅěẻẹęȩḝǝĭȋîĩḭï'..
			'ḯīíìȉǐỉịįıŏȏôỗốồổộõṏṍöōṑóṓòȍǒỏọǫǭơỡớờởợøǿŭȗûṷũṻṹṵüǖǘǜǚṳūúùȕǔủůụųưữứừửựŷỹÿȳýỳỷẙỵ'
		str = mw.ustring.lower(mw.ustring.sub(str,1,1))
		return mw.ustring.find(vowels, str, 1, true ) 
	end
end

------------------------------------------------------------------------------------------
function p.inparentheses(str)
	if (not str) or (str == "") then
		return nil
	end
	local str2 = mw.message.new('parentheses', str):inLanguage(lang):plain()
	return (str2~=nil and str2) or ('('..str..')') -- in case this module is moved to a project where {{int:parenthesis}} is not set
end

------------------------------------------------------------------------------------------
function p.of(word, lang, raw, gender, number, determiner) -- rough translation of "of" in various languages
-- note that the cases when on "of" is employed varies a lot among languages, so it is more prudent to call this from lang specific function only
	if not raw or mw.text.trim(raw) == "" then 
		raw = mw.ustring.lower(nowiki(word))
	end

	-- raw is the string without the Wikiformatting so that it correctly analyses the string that is [[:fr:Italie|Italie]] -> 'italie'
	-- any way to automate this ?
	
	if lang == 'fr' then 
		determiner = yesno(determiner,false)

		if determiner then
			if string.sub(number or '',1,1)=='p' then -- number == 'plural'
				return 'des&#32;' .. word
			elseif p.vowelfirst(raw) then
				return 'de l’' .. word
			elseif string.sub(gender or '',1,1)=='f' then -- gender == 'feminine'
				return 'de la&#32;' .. word
			else
				return 'du&#32;' .. word
			end
		else
			if p.vowelfirst(raw) then
				return 'd’' .. word
			else
				return 'de&#32;' .. word
			end
		end
	elseif lang == 'ca' then
		-- implement [[Template:Of/ca]] or  https://ca.wikipedia.org/wiki/Plantilla:Deod%27/base for case where "{{{context}}}" is "en" (default on Commons)
		if ( p.vowelfirst(raw) and not mw.ustring.find( 'ia|ià|ie|io|iu|ua|ue|ui|uí|uï|uo|ya|ye|yi|yo|yu|', mw.ustring.sub(raw,1,2) .. '|')) then
			return 'd\'' .. word
		else
			return 'de ' .. word
		end
	end	
end

------------------------------------------------------------------------------------------
function p.noungroup(noun, adj, lang)
	if not noun or noun == '' then 
		return nil -- not '' so that it is not counted as a string by mw.listToText
	end
	if not adj or adj == ''
		then return noun
	end
	
	local wordsep = mw.message.new( "Word-separator" ):inLanguage(lang):plain()
	-- Assign order of words per language: 1 means adjective before the noun and 2 means adjective after the noun
	-- original source for many: language subpages of [[Template:Technique]]
	-- corrections to that using https://wals.info/feature/87A, still different:
	-- * vi: WALS says 2
	-- * gl isn't consistent in [[Template:Technique/gl]], but seems to be 2
	local LUT = {ar=2, ca=2, cs=1, da=1, de=1, el=1, en=1, es=2, et=1, fi=1, fr=2, 
		gl=2, he=2, hu=1, it=2, ja=1, la=2, mk=1, nds=1, nb=1, nl=1, no=1, pl=1, pt=2, 
		ro=2, ru=1, scn=2, sk=1, sl=1, sr=1, sv=1, sw=1, tr=1, vec=1, vi=1, zh=1}
	local primaryLangSubtag = mw.text.split(lang, '-', true)[1]
	local case = LUT[lang] or LUT[primaryLangSubtag]
	if case==1 then     -- adjective before the noun
		return adj .. wordsep .. noun
	elseif case==2 then -- adjective after the noun
		return noun .. wordsep .. adj
	else                -- order unknown
		return noun .. ' (' .. adj .. ')'
	end
end

------------------------------------------------------------------------------------------
function p.conj(args, lang, conjtype)
	local comma    = mw.message.new( "comma-separator"):inLanguage(lang):plain()
	local wordsep  = mw.message.new( "Word-separator" ):inLanguage(lang):plain()
	local andtable = { ar=' و', he=' ו', ja='および', pl=' i '} -- languages with a problem with the MediaWiki:And
	local sep1, sep2 = comma, nil
	if conjtype == 'comma' then
		sep2 = comma
	elseif conjtype == 'or' or conjtype == 'explicit or' then 
		local wordor = formatMessage('I18n/Or.tab', 'or', lang) -- get translated "or"
		sep2 = wordsep .. wordor  .. wordsep -- pad it with spaces (if needed)
		sep1 = (conjtype == 'explicit or' and sep2) or comma -- adds "or" betwen all words when the context can be confusing
	elseif conjtype and conjtype ~= 'and' and conjtype ~= '' then
		sep1, sep2 = conjtype, conjtype
	elseif andtable[lang] then 
		sep2 = andtable[lang]
	else
		sep2 = mw.message.new( "and" ):inLanguage(lang):plain() .. wordsep
	end
	return mw.text.listToText(args, sep1, sep2)

end

-- ===========================================================================
-- === Version of the functions to be called from template namespace
-- ===========================================================================

------------------------------------------------------------------------------------------
function p.offromwiki(frame)
	local args = frame.args
	return p.of(args.word, args.lang, args.raw, args.gender, args.number, args.determiner)
end

------------------------------------------------------------------------------------------
function p.conjfromWiki(frame)
	local args = frame.args
	if not args or not args[1] then
		args = frame:getParent().args
	end
	local lang = args.lang
	if not lang or mw.text.trim(lang) == '' then
		lang = frame:callParserFunction( "int", "lang" )
	end
	local newargs = {}  -- transform args metatable into a table so it can be concetenated
	for i, j in pairs(args) do
			if type(i) == 'number' then
				j = mw.text.trim(j)
				if j ~= '' then
					table.insert(newargs, j)
				end
			else 
				if i ~= 'type' and i ~= 'lang' then 
					return 'error: bad parameter in template:Conj: ' .. i .. '[[Category:Pages with incorrect template usage/Conj|A]]'
				end
			end
	end
	return p.conj(newargs, lang, args.type)
end

return p