Module:A or an

Wikipedia Content This page was copied from Wikipedia. The original source may be accessed here. This Wikipedia content is available under the Creative Commons Attribution ShareAlike 3.0 Unported License.

This module implements Template:a or an. Exception words are stored at Module:a or an/words.

---

local p = {}
local words = mw.loadData('Module:A or an/words')

local lcVChars = 'aeiouà-æè-ïò-öø-üāăąēĕėęěĩīĭįıijōŏőœũūŭůűų'
local ucVvChars = 'AEFHILMNORSXÀ-ÆÈ-ÏÒ-ÖØĀĂĄĒĔĖĘĚĨĪĬĮıIJŌŎŐŒÑĤĦĹĻĽĿŁŃŅŇŊŔŖŘŚŜŞ'

local function findWord(s, t)
	for i, v in ipairs(t) do
		if mw.ustring.find(s, '^' .. v .. '$') then
			return true
		end
	end
end

function p._main(args)
	local s = args[1] and mw.text.trim(args[1])
	local pron = 'a'
	local ret = ''
	
	if s and s ~= '' then
		local origStr = s
		
		s = mw.ustring.gsub(s, '</?[A-Za-z][^>]->', '') -- Remove HTML tags
		s = mw.ustring.gsub(s, '%[%[[^%|]+%|(..-)%]%]', '%1') -- Remove wikilinks
		s = mw.ustring.gsub(mw.ustring.gsub(s, '%[%[', ''), '%]%]', '')
		s = mw.ustring.gsub(s, '^["%$\'%(<%[%{¢-¥₠-₿]+', '') -- Strip some symbols at the beginning
		s = mw.ustring.match(s, '^%.?[0-9%u%l]+') or s -- Extract the first word
		
		if mw.ustring.find(s, '^[0-9]') then -- It begins with a number
			s = mw.ustring.match(s, '^[0-9]+') -- Extract the number
			if findWord(s, words.vNums) then -- '18' etc.
				pron = 'an'
			end
		elseif mw.ustring.match(s, '^[0-9%u]+$') then -- It looks like an acronym
			if mw.ustring.find(s, '^[' .. ucVvChars .. ']')
				and not findWord(s, words.cvAcronyms) -- Exclude 'NASA' etc.
				or findWord(s, words.vvAcronyms) -- 'UNRWA' etc.
			then
				pron = 'an'
			end
		else
			s = mw.ustring.lower(s) -- Uncapitalize
			if mw.ustring.find(s, '^['.. lcVChars .. ']') then -- It begins with a vowel
				if not findWord(s, words.vcWords) -- Exclude 'euro' etc.
					or findWord(s, words.vvWords) -- But not 'Euler' etc.
				then
					pron = 'an'
				end
			elseif args.variety and mw.ustring.lower(args.variety) == 'us' -- 'herb' etc.
				and findWord(s, words.cvWordsUS)
				or findWord(s, words.cvWords) -- 'hour' etc.
			then
				pron = 'an'
			end
		end
		ret = pron .. ' ' .. origStr
	end
	
	return ret
end

function p.main(frame)
	return p._main(frame:getParent().args)
end

return p