local U = mw.ustring.char
local rsub = mw.ustring.gsub -- WARNING: Don't return this directly in a function, or surround in parens
local export = {}
local fatHatan = U(0x64B) -- اً, tanvin-e nasb (تنوین نصب)
local fathe = U(0x64E) -- also zebar, zabar
local kasre = U(0x650) -- also zir
local zamme = U(0x64F) -- also piš
local tashdid = U(0x651) -- also called šadda
local jazm = U(0x652)
local alef = "ا" -- Arabic alif
local zwnj = U(0x200C)
local he = "ه"
local hamza_above = U(0x0654)
local xva = "خوا"
local xvi = "خوی"
local eyi = "های"
local vaw = U(0x0648)
local ye = U(0x06CC)
local group = "بپتثجچحخدذرزژسشصضطظغفقکگلمنوهی"
local ZZP = "َُِ"
local mapping = {
["ا"] = 'â', ["ب"] = 'b', ["پ"] = 'p', ["ت"] = 't', ["ث"] = 's', ["ج"] = 'j', ["چ"] = 'č', ["ح"] = 'h', ["خ"] = 'x',
["د"] = 'd', ["ذ"] = 'z', ["ر"] = 'r', ["ز"] = 'z', ["ژ"] = 'ž', ["س"] = 's', ["ش"] = 'š', ["ص"] = 's', ["ض"] = 'z',
["ط"] = 't', ["ظ"] = 'z', ["غ"] = 'ğ', ["ف"] = 'f', ["ق"] = 'q', ["ک"] = 'k', ["گ"] = 'g', ["ل"] = 'l',
["م"] = 'm', ["ن"] = 'n', ["و"] = 'u', ["ه"] = 'h', ["ی"] = 'i', ["آ"] = 'â',
-- displaying on separate lines as the viewing becomes distorted on these combinations
["ع"] = "'",
["ء"] = "'",
["ئ"] = "'",
["ؤ"] = "'",
["أ"] = "'",
-- diacritics
[fathe] = "a",
[kasre] = "e",
[zamme] = "o",
[jazm] = "", -- also sokun - no vowel
[fatHatan] = "n",
[zwnj] = "-",
["ۀ"] = "-ye", -- U+06C0 ARABIC LETTER HEH WITH YEH ABOVE (non-normative; should use the sequence U+0647 U+0654 = هٔ)
[xva] = "xâ",
[xvi] = "xi",
[eyi] = "e-yi",
-- ligatures
["ﻻ"] = "lâ",
["ﷲ"] = "llâh",
-- kashida (= Arabic tatweel)
["ـ"] = "", -- kashida, no sound
-- numerals
["۱"] = "1", ["۲"] = "2", ["۳"] = "3", ["۴"] = "4", ["۵"] = "5",
["۶"] = "6", ["۷"] = "7", ["۸"] = "8", ["۹"] = "9", ["۰"] = "0",
-- normal arabic variants to numerals
["١"] = "1", ["٢"] = "2", ["٣"] = "3", ["٤"] = "4", ["٥"] = "5",
["٦"] = "6", ["٧"] = "7", ["٨"] = "8", ["٩"] = "9", ["٠"] = "0",
-- punctuation (leave on separate lines)
["؟"] = "?", -- question mark
["،"] = ",", -- comma
["؛"] = ";", -- semicolon
["«"] = '“', -- quotation mark
["»"] = '”', -- quotation mark
["٪"] = "%", -- percent
["؉"] = "‰", -- per mille
["٫"] = ".", -- decimals
["٬"] = ",", -- thousand
}
function export.tr(text, lang, sc)
text = rsub(text, '([' .. group .. ']' .. tashdid .. '?)ه$', '%1e')
text = rsub(text, alef .. fathe, "a")
text = rsub(text, he .. zwnj, "e-")
text = rsub(text, he .. ye, "-ye")
text = rsub(text, "هٔ", "-ye")
text = rsub(text, xva, "xâ")
text = rsub(text, kasre .. ye .. '([' .. group .. '])', "ey%1")
text = rsub(text, jazm .. alef, "a")
--text = rsub(text, 'ىٰ', "â") -- the first letter is U+0649 (Arabic alif maqṣūra), it doesn't belong here
text = rsub(text, 'یٰ', "â") -- the first letter is U+06CC (Farsi ye)
text = rsub(text, 'ٰ', "â")
text = rsub(text, 'ا' .. fatHatan, "an")
-- text = rsub(text, 'الله', "ﷲ")
-- text = rsub(text, 'لا', "ﻻ")
--- ezafe
text = rsub(text, '([' .. group .. '])' .. kasre .. " ", "%1-e ")
text = rsub(text, '.', mapping)
text = rsub(text, 'aâ', "â")
text = rsub(text, 'âa', "a")
text = rsub(text, 'u([aâeiou])', "v%1")
text = rsub(text, 'i([aâeiou])', "y%1")
text = rsub(text, "([aâeiou])(" .. tashdid .. ")", "%2%1") -- swapping tašdid with vowels
text = rsub(text, "(.)" .. tashdid, "%1%1") -- implementing tašdid
text = rsub(text, 'eh$', "e")
text = rsub(text, "eh([^aâeiouy'bdfghjklmnpqrstvyxzčğšž])", "e%1")
return text
end
return export
