La documentation pour ce module peut être créée à Module:HiddenUnicode/doc
require('strict') local char = mw.ustring.char local find = string.find -- Format control characters: these are multibyte characters in UTF-8 (not part of ASCII). -- Unfortunately, Lua patterns don't support alternations like in regular expressions, and -- can't match non-ASCII characters using 1-byte character classes). -- So we need to try matching each one using a loop. local hiddenUnicodeChars = { -- Bidi controls, strongly discouraged in HTML (better use "bdi" or "bdo" elements) char(0x200E), -- LEFT-TO-RIGHT MARK (LRM) char(0x200F), -- RIGHT-TO-LEFT MARK (RLM) char(0x202A), -- LEFT-TO-RIGHT EMBEDDING (LRE) char(0x202B), -- RIGHT-TO-LEFT EMBEDDING (RLE) char(0x202C), -- POP DIRECTIONAL FORMATTING (PDF) char(0x202D), -- LEFT-TO-RIGHT OVERRIDE (LRO) char(0x202E), -- RIGHT-TO-LEFT OVERRIDE (RLO) char(0x2066), -- LEFT-TO-RIGHT ISOLATE (LRI) char(0x2067), -- RIGHT-TO-LEFT ISOLATE (RLI) char(0x2068), -- FIRST STRONG ISOLATE (FSI) char(0x2069), -- POP DIRECTIONAL ISOLATE (PDI) -- Invisible zero-width semantic characters char(0x200B), -- ZERO-WIDTH SPACE char(0x2060), -- WORD JOINER (WJ) -- Invisible zero-width mathematical characters (better use math elements in HTML) char(0x2061), -- FUNCTION APPLICATION (contiguity operator indicating function application) char(0x2062), -- INVISIBLE TIMES (contiguity operator indicating multiplication) char(0x2063), -- INVISIBLE SEPARATOR (invisible comma, contiguity operator indicating that adjacent mathematical symbols form a list, e.g. when no visible comma is used between multiple indices) char(0x2064), -- INVISIBLE PLUS (contiguity operator indicating addition) -- Deprecated, use of these characters is strongly discouraged in all texts (not only in HTML). char(0x206A), -- INHIBIT SYMMETRIC SWAPPING char(0x206B), -- ACTIVATE SYMMETRIC SWAPPING char(0x206C), -- INHIBIT ARABIC FORM SHAPING char(0x206D), -- ACTIVATE ARABIC FORM SHAPING char(0x206E), -- NATIONAL DIGIT SHAPES char(0x206F), -- NOMINAL DIGIT SHAPES char(0xFEFF), -- ZERO WIDTH NO-BREAK SPACE (ZWNBSP) (deprecated in favor of WJ, now used only for byte order marks) } local p = {} function p.HiddenUnicode(frame) local t = frame.args[1] or '' for _, c in ipairs(hiddenUnicodeChars) do if t:find(c, 1, true) then return '[[Catégorie:Listing avec un caractère unicode caché]]' end end return '' end return p