Module:ar-translit

Hello, you have come here looking for the meaning of the word Module:ar-translit. In DICTIOUS you will not only get to know all the dictionary meanings for the word Module:ar-translit, but we will also tell you about its etymology, its characteristics and you will know how to say Module:ar-translit in singular and plural. Everything you need to know about the word Module:ar-translit you have here. The definition of the word Module:ar-translit will help you to be more precise and correct when speaking or writing your texts. Knowing the definition ofModule:ar-translit, as well as those of other words, enriches your vocabulary and provides you with more and better linguistic resources.
The following documentation is located at Module:ar-translit/documentation. Categories were auto-generated by Module:module categorization.
Useful links: subpage list • links • transclusions • testcases • sandbox (diff)

This module will transliterate Arabic language text per WT:AR TR. It is also used to transliterate Adyghe, Avar, Chechen, Ingush, and Kabardian. The module should preferably not be called directly from templates or other modules. To use it from a template, use {{xlit}}. Within a module, use Module:languages#Language:transliterate.

For testcases, see Module:ar-translit/testcases.

Functions

tr(text, lang, sc)
Transliterates a given piece of text written in the script specified by the code sc, and language specified by the code lang.
When the transliteration fails, returns nil.

-- Authors: Benwing, ZxxZxxZ, Atitarev

local export = {}

local m_str_utils = require("Module:string utilities")

local gcodepoint = m_str_utils.gcodepoint
local rfind = m_str_utils.find
local rsubn = m_str_utils.gsub
local rmatch = m_str_utils.match
local rsplit = m_str_utils.split
local U = m_str_utils.char

-- assigned below
local has_diacritics

-- version of rsubn() that discards all but the first return value
local function rsub(term, foo, bar)
	local retval = rsubn(term, foo, bar)
	return retval
end

local zwnj = U(0x200C) -- zero-width non-joiner
local alif_madda = U(0x622)
local alif_hamza_below = U(0x625)
local alif = U(0x627)
local taa_marbuuTa = U(0x629)
local laam = U(0x644)
local waaw = U(0x648)
local alif_maqSuura = U(0x649)
local yaa = U(0x64A)
local fatHataan = U(0x64B)
local Dammataan = U(0x64C)
local kasrataan = U(0x64D)
local fatHa = U(0x64E)
local Damma = U(0x64F)
local kasra = U(0x650)
local shadda = U(0x651)
local sukuun = U(0x652)
local dagger_alif = U(0x670)
local alif_waSl = U(0x671)
--local zwj = U(0x200D) -- zero-width joiner
local lrm = U(0x200E) -- left-to-right mark
local rlm = U(0x200F) -- right-to-left mark
-- Occurs after al- in allaḏī and variants so that we can implement elision of
-- a- after a preceding vowel, after which we remove the marker.
local alladi_marker = U(0xFFF0)

local tt = {
	-- consonants
	="b", ="t", ="ṯ", ="j", ="ḥ", ="ḵ",
	="d", ="ḏ", ="r", ="z", ="s", ="š",
	="ṣ", ="ḍ", ="ṭ", ="ẓ", ="ʕ", ="ḡ",
	="f", ="q", ="k", ="k", ="l", ="m", ="n",
	="h",
	-- tāʾ marbūṭa (special) - always after a fátḥa (a), silent at the end of
	-- an utterance, "t" in ʾiḍāfa or with pronounced tanwīn. We catch
	-- most instances of tāʾ marbūṭa before we get to this stage.
	="t", -- tāʾ marbūṭa = ة
	-- control characters
	="-", -- ZWNJ (zero-width non-joiner)
	-- ="", -- ZWJ (zero-width joiner)
	-- rare letters
	="p", ="č", ="ž", ="v", ="v", ="g",
	="g", ="q", ="f", ="n", ="g",
	-- semivowels or long vowels, alif, hamza, special letters
	="ā", -- ʾalif
	-- hamzated letters
	="ʔ", -- hamza over alif
	="ʔ", -- hamza under alif
	="ʔ", -- hamza over wāw
	="ʔ", -- hamza over yā
	="ʔ", -- hamza on the line
	-- long vowels
	="w", --"ū" after ḍamma (u) and not before diacritic
	="y", --"ī" after kasra (i) and not before diacritic
	="ā", -- ʾalif maqṣūra
	="ʔā", -- ʾalif madda
	= "", -- hamzatu l-waṣl
	 = "ā", -- ʾalif xanjariyya = dagger ʾalif (Koranic diacritic)
	-- short vowels, šádda and sukūn
	="an", -- fatḥatan
	="un", -- ḍammatan
	="in", -- kasratan
	="a", -- fatḥa
	="u", -- ḍamma
	="i", -- kasra
	-- šadda - doubled consonant
	="", --sukūn - no vowel
	-- ligatures
	="lā",
	="llāh",
	-- taṭwīl
	="", -- taṭwīl, no sound
	-- numerals
	="1", ="2", ="3", ="4", ="5",
	="6", ="7", ="8", ="9", ="0",
	-- punctuation (leave on separate lines)
	="?", -- question mark
	='“', -- quotation mark
	='”', -- quotation mark
	=".", -- decimal point
	=",", -- thousands separator
	="%", -- percent sign
	=",", -- comma
	=";" -- semicolon
}

local sun_letters = "تثدذرزسشصضطظلن"
-- For use in implementing sun-letter assimilation of ال (al-)
local ttsun1 = {}
local ttsun2 = {}
local ttsun3 = {}
for cp in gcodepoint(sun_letters) do
	local ch = U(cp)
	ttsun1 = tt
	ttsun2 = tt .. "-" .. ch
	table.insert(ttsun3, tt)
end
-- For use in implementing elision of al-
local sun_letters_tr = table.concat(ttsun3, "")

local consonants_needing_vowels = "بتثجحخدذرزسشصضطظعغفقكڪلمنهپچژڤگڨڧڢںڭأإؤئءةﷲ"
-- consonants on the right side; includes alif madda
local rconsonants = consonants_needing_vowels .. "ويآ"
-- consonants on the left side; does not include alif madda
local lconsonants = consonants_needing_vowels .. "وي"
-- Arabic semicolon, comma, question mark; taṭwīl; period, exclamation point,
-- single quote for bold/italic, double quotes for quoted material
local punctuation = "؟،؛" .. "ـ" .. ".!'" .. '"'
local space_like = "%s'" .. '"'
local space_like_class = ""
local numbers = "١٢٣٤٥٦٧٨٩٠"

local before_diacritic_checking_subs = {
	------------ transformations prior to checking for diacritics --------------
	-- random Koranic marks and presentation forms
	{U(0x06E1), sukuun}, -- "Small High Dotless Head of Khah" (variant of sukūn)
	{U(0x06DA), ""}, -- "Small High Jeem"
	{U(0x06DF), ""}, -- "Small High Rounded Zero" (FIXME: correct?)
	{U(0x08F0), U(0x64B)}, -- "Open Fathatan"
	{U(0x08F1), U(0x64C)}, -- "Open Dammatan"
	{U(0x08F2), U(0x64D)}, -- "Open Kasratan"
	{U(0x06E4), ""}, -- "Small High Madda" (FIXME: correct?)
	{U(0x06D6), ""}, -- "Small High Ligature Sad with Lam with Alef Maksura" (FIXME: there are others we need to do)
	{U(0x06E5), "و"},
	{U(0x06E6), "ي"},
	-- convert llh for allāh into ll+shadda+dagger-alif+h
	{"لله", "للّٰه"},
	-- shadda+short-vowel (including tanwīn vowels, i.e. -an -in -un) gets
	-- replaced with short-vowel+shadda during NFC normalisation, which
	-- MediaWiki does for all Unicode strings; however, it makes the
	-- transliteration process inconvenient, so undo it.
	{"()" .. shadda, shadda .. "%1"},
	-- ignore Koranic gemination at beginning of word due to assimilation of preceding consonant
	{" ()" .. shadda, " %1"},
	-- ignore alif jamīla (otiose alif in 3pl verb forms)
	--     #1: handle ḍamma + wāw + alif (final -ū)
	{Damma .. waaw .. alif, Damma .. waaw},
	--     #2: handle wāw + sukūn + alif (final -w in -aw in defective verbs)
	--     this must go before the generation of w, which removes the waw here.
	{waaw .. sukuun .. alif, waaw .. sukuun},
	-- ignore final alif or alif maqṣūra following fatḥatan (e.g. in accusative
	-- singular or words like عَصًا "stick" or هُدًى "guidance"; this is called
	-- tanwin nasb)
	{fatHataan .. "", fatHataan},
	-- same but with the fatḥatan placed over the alif or alif maqṣūra
	-- instead of over the previous letter (considered a misspelling but
	-- common)
	{"" .. fatHataan, fatHataan},
	-- tāʾ marbūṭa should always be preceded by fatḥa, alif, alif madda or
	-- dagger alif; infer fatḥa if not
	{"()" .. taa_marbuuTa, "%1" .. fatHa .. taa_marbuuTa},
	-- similarly for alif between consonants, possibly marked with shadda
	-- (does not apply to initial alif, which is silent when not marked with
	-- hamza, or final alif, which might be pronounced as -an)
	{"(" .. shadda .. "?)" .. alif .. "()",
		"%1" .. fatHa .. alif .. "%2"},
	-- infer fatḥa in case of non-fatḥa + alif/alif-maqṣūra + dagger alif
	{"()(" .. dagger_alif .. ")", "%1" .. fatHa .. "%2"},
	-- infer kasra in case of hamza-under-alif not + kasra
	{alif_hamza_below .. "()", alif_hamza_below .. kasra .. "%1"},
	-- ignore dagger alif placed over regular alif or alif maqṣūra
	{"()" .. dagger_alif, "%1"},

	----------- rest of these concern definite article alif-lām ----------
	-- in kasra/ḍamma + alif + lam, make alif into hamzatu l-waṣl, so we
	-- handle cases like بِالتَّوْفِيق (bi-t-tawfīq) correctly
	{"()" .. alif .. laam, "%1" .. alif_waSl .. laam},
	-- al + consonant + shadda (only recognize word-initially if regular alif): remove shadda
	{"^(" .. alif .. fatHa .. "?" .. laam .. ")" .. shadda, "%1"},
	{"(" .. space_like_class .. alif .. fatHa .. "?" .. laam .. ")" .. shadda, "%1"},
	{"(" .. alif_waSl .. fatHa .. "?" .. laam .. ")" .. shadda, "%1"},
	-- handle l- hamzatu l-waṣl or word-initial al-
	{"^" .. alif .. fatHa .. "?" .. laam, "al-"},
	{"(" .. space_like_class .. ")" .. alif .. fatHa .. "?" .. laam, "%1al-"},
	-- next one for bi-t-tawfīq
	{"()" .. alif_waSl .. fatHa .. "?" .. laam, "%1-l-"},
	-- next one for remaining hamzatu l-waṣl (at beginning of word)
	{alif_waSl .. fatHa .. "?" .. laam, "l-"},
	-- special casing if the l in al- has a shadda on it (as in الَّذِي "that"),
	-- so we don't mistakenly double the dash; insert a special marker here so
	-- that we know later to elide the a- after a vowel
	{"l%-" .. shadda, "l" .. alladi_marker .. "l"},
	-- implement assimilation of sun letters
	{"l%-", ttsun2},
}

-- Transliterate the word(s) in TEXT. LANG (the language) and SC (the script)
-- are ignored. OMIT_I3RAAB means leave out final short vowels (ʾiʿrāb).
-- GRAY_I3RAAB means render transliterate short vowels (ʾiʿrāb) in gray.
-- FORCE_TRANSLIT causes even non-vocalized text to be transliterated
-- (normally the function checks for non-vocalized text and returns nil,
-- since such text is ambiguous in transliteration).
function export.tr(text, lang, sc, omit_i3raab, gray_i3raab, force_translit)
	-- make it possible to call this function from a template
	if type(text) == "table" then
		local function f(x) return (x ~= "") and x or nil end
		text, lang, sc, omit_i3raab, force_translit =
			f(text.args), f(text.args), f(text.args), f(text.args), f(text.args)
	end

	for _, sub in ipairs(before_diacritic_checking_subs) do
		text = rsub(text, sub, sub)
	end

	if not force_translit and not has_diacritics(text) then
		require("Module:debug").track("ar-translit/lacking diacritics")
		return nil
	end
	
	------------ transformations after checking for diacritics --------------
	-- Replace plain alif with hamzatu l-waṣl when followed by fatḥa/ḍamma/kasra.
	-- Must go after handling of initial al-, which distinguishes alif-fatḥa
	-- from alif w/hamzatu l-waṣl. Must go before generation of ū and ī, which
	-- eliminate the ḍamma/kasra.
	text = rsub(text, alif .. "()", alif_waSl .. "%1")
	-- ḍamma + waw not followed by a diacritic is ū, otherwise w
	text = rsub(text, Damma .. waaw .. "()", "ū%1")
	text = rsub(text, Damma .. waaw .. "$", "ū")
	-- kasra + yaa not followed by a diacritic (or ū from prev step) is ī, otherwise y
	text = rsub(text, kasra .. yaa .. "()", "ī%1")
	text = rsub(text, kasra .. yaa .. "$", "ī")
	-- convert shadda to double letter.
	text = rsub(text, "(.)" .. shadda, "%1%1")
	if not omit_i3raab and gray_i3raab then -- show ʾiʿrāb grayed in transliteration
		-- decide whether to gray out the t in ﺓ. If word begins with al- or l-, yes.
		-- Otherwise, no if word ends in a/i/u, yes if ends in an/in/un.
		text = rsub(text, "^(a?l%-+)" .. taa_marbuuTa .. "()",
			'%1<span style="color: #888888">t</span>%2')
		text = rsub(text, "(" .. space_like_class .. "a?l%-+)" .. taa_marbuuTa .. "()",
			'%1<span style="color: #888888">t</span>%2')
		text = rsub(text, taa_marbuuTa .. "()", "t%1")
		text = rsub(text, taa_marbuuTa .. "()",
			'<span style="color: #888888">t</span>%1')
		text = rsub(text, ".", {
			 = '<span style="color: #888888">an</span>',
			 = '<span style="color: #888888">in</span>',
			 = '<span style="color: #888888">un</span>'
		})
		text = rsub(text, "()(" .. space_like_class .. ")",
			function(vowel, space)
				vowel_repl = {
					 = '<span style="color: #888888">a</span> ',
					 = '<span style="color: #888888">i</span> ',
					 = '<span style="color: #888888">u</span> '
				}
				return vowel_repl .. space
			end
		)
		text = rsub(text, "$", {
			 = '<span style="color: #888888">a</span>',
			 = '<span style="color: #888888">i</span>',
			 = '<span style="color: #888888">u</span>'
		})
		text = rsub(text, '</span><span style="color: #888888">', "")
	elseif omit_i3raab then -- omit ʾiʿrāb in transliteration
		text = rsub(text, "", "")
		text = rsub(text, "(" .. space_like_class .. ")", "%1")
		text = rsub(text, "$", "")
	end
	-- tāʾ marbūṭa should not be rendered by -t if word-final even when
	-- ʾiʿrāb (desinential inflection) is shown; instead, use (t) before
	-- whitespace, nothing when final; but render final -ﺍﺓ and -ﺁﺓ as -āh,
	-- consistent with Wehr's dictionary
	-- Left-to-right or right-to-left mark at end of text will prevent tāʾ marbūṭa
	-- from being transliterated correctly.
	text = string.gsub(text, lrm, "")
	text = string.gsub(text, rlm, "")
	text = rsub(text, "()" .. taa_marbuuTa .. "$", "%1h")
	-- Ignore final tāʾ marbūṭa (it appears as "a" due to the preceding
	-- short vowel). Need to do this after graying or omitting word-final
	-- ʾiʿrāb.
	text = rsub(text, taa_marbuuTa .. "$", "")
	text = rsub(text, taa_marbuuTa .. "(%p)", "%1")
	if not omit_i3raab then -- show ʾiʿrāb in transliteration
		text = rsub(text, taa_marbuuTa .. "(" .. space_like_class .. ")", "(t)%1")
	else
		-- When omitting ʾiʿrāb, show all non-absolutely-final instances of
		-- tāʾ marbūṭa as (t), with trailing ʾiʿrāb omitted.
		text = rsub(text, taa_marbuuTa, "(t)")
	end
	-- tatwīl should be rendered as - at beginning or end of word. It will
	-- be rendered as nothing in the middle of a word (FIXME, do we want
	-- this?)
	text = rsub(text, "^ـ", "-")
	text = rsub(text, "(" .. space_like_class .. ")ـ",
		"%1-")
	text = rsub(text, "ـ$", "-")
	text = rsub(text, "ـ(" .. space_like_class .. ")", "-%1")
	-- Now convert remaining Arabic chars according to table.
	text = rsub(text, ".", tt)
	text = rsub(text, "aā", "ā")
	-- Implement elision of al- after a final vowel. We do this
	-- conservatively, only handling elision of the definite article and related
	-- terms (specifically, relative pronoun الَّذِي (allaḏī) and variants) rather
	-- than elision in other cases of hamzat al-waṣl (e.g. form-I imperatives
	-- or form-VII and above verbal nouns) partly because elision in
	-- these cases isn't so common in MSA and partly to avoid excessive
	-- elision in case of words written with initial bare alif instead of
	-- properly with hamzated alif. Possibly we should reconsider.
	text = rsub(text, "('* +'*)a()",
		"%1%2")
	if gray_i3raab then
		text = rsub(text, "('*</span>'* +'*)a()",
			"%1%2")
	end
	-- remove indicator of allaḏī, which has served its purpose
	text = rsub(text, alladi_marker, "")
	-- Special-case the transliteration of allāh, without the hyphen.
	text = rsub(text, "^(a?)l%-lāh", "%1llāh")
	text = rsub(text, "(" .. space_like_class .. "a?)l%-lāh", "%1llāh")
	-- Compress multiple spaces, which may occur e.g. when removing Koranic diacritics.
	text = rsub(text, "(%s)%s+", "%1")

	return text
end

local has_diacritics_subs = {
	-- FIXME! What about lam-alif ligature?
	-- remove punctuation and shadda
	-- must go before removing final consonants
	{"", ""},
	-- Remove consonants at end of word or utterance, so that we're OK with
	-- words lacking iʿrāb (must go before removing other consonants).
	-- If you want to catch places without iʿrāb, comment out the next two lines.
	{"$", ""},
	{"(}]?" .. space_like_class .. ")", "%1"},
	-- remove consonants (or alif) when followed by diacritics
	-- must go after removing shadda
	-- do not remove the diacritics yet because we need them to handle
	-- long-vowel sequences of diacritic + pseudo-consonant
	{"()", "%1"},
	-- the following two must go after removing consonants w/diacritics because
	-- we only want to treat vocalic wāw/yā' in them (we want to have removed
	-- wāw/yā' followed by a diacritic)
	-- remove ḍamma + wāw
	{Damma .. waaw, ""},
	-- remove kasra + yā'
	{kasra .. yaa, ""},
	-- remove fatḥa/fatḥatan + alif/alif-maqṣūra
	{"", ""},
	-- remove diacritics
	{"", ""},
	-- remove numbers, hamzatu l-waṣl, alif madda
	{"", ""},
	-- remove non-Arabic characters
	{"[^" .. U(0x0600) .. "-" .. U(0x06FF) .. U(0x0750) .. "-" .. U(0x077F) ..
			 U(0x08A0) .. "-" .. U(0x08FF) .. U(0xFB50) .. "-" .. U(0xFDFF) ..
			 U(0xFE70) .. "-" .. U(0xFEFF) .. "]", ""}
}

-- declared as local above
function has_diacritics(text)
	local orig_text = text
	local count
	text, count = rsubn(text, "", "")
	if count > 0 then
		require("Module:debug").track("ar-translit/lrm or rlm")
	end
	for _, sub in ipairs(has_diacritics_subs) do
		text = rsub(text, unpack(sub))
	end
	if #text > 0 then
		mw.log(("Check for missing diacritics failed; original text '%s', text without diacritics '%s'"):format(
			orig_text, text))
	end
	return #text == 0
end

-- Return true if transliteration TR is an irregular transliteration of
-- ARABIC. Return false if ARABIC can't be transliterated. For purposes of
-- establishing regularity, hyphens are ignored and word-final tāʾ marbūṭa
-- can be transliterated as "(t)", "" or "t".
function export.irregular_translit(arabic, tr)
	if not arabic or arabic == "" or not tr or tr == "" then
		return false
	end
	local regtr = export.tr(arabic)
	if not regtr or regtr == tr then
		return false
	end
	local arwords = rsplit(arabic, " ")
	local regwords = rsplit(regtr, " ")
	local words = rsplit(tr, " ")
	if #regwords ~= #words or #regwords ~= #arwords then
		return true
	end
	for i=1,#regwords do
		local regword = regwords
		local word = words
		local arword = arwords
		-- Resolve final (t) in auto-translit to t, h or nothing
		if rfind(regword, "%(t%)$") then
			regword = rfind(word, "āh$") and rsub(regword, "%(t%)$", "h") or
				rfind(word, "t$") and rsub(regword, "%(t%)$", "t") or
				rsub(regword, "%(t%)$", "")
		end
		-- Resolve clitics + short a + alif-lām, which may get auto-transliterated
		-- to contain long ā, to short a if the manual translit has it; note
		-- that currently in cases with assimilated l, the auto-translit will
		-- fail, so we won't ever get here and don't have to worry about
		-- auto-translit l against manual-translit assimilated char.
		local clitic_chars = "^" -- separate line to avoid L2R display weirdness
		if rfind(arword, clitic_chars .. fatHa .. "?" .. laam) and rfind(word, "^a%-") then
			regword = rsub(regword, "^()ā", "%1a")
		end
		-- Ignore hyphens when comparing
		if rsub(regword, "%-", "") ~= rsub(word, "%-", "") then
			return true
		end
	end
	return false
end

return export
Wikious
Boobota
Sagapedia