From 0d47952611198ef6b1163f366dc03922d20b1475 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Wed, 17 Apr 2024 09:42:04 +0200 Subject: Adding upstream version 7.94+git20230807.3be01efb1+dfsg. Signed-off-by: Daniel Baumann --- nselib/re.lua | 321 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 321 insertions(+) create mode 100644 nselib/re.lua (limited to 'nselib/re.lua') diff --git a/nselib/re.lua b/nselib/re.lua new file mode 100644 index 0000000..43e78cf --- /dev/null +++ b/nselib/re.lua @@ -0,0 +1,321 @@ +--- +-- Regular Expression functions +-- +-- This is the re.lua module included in the LPeg distribution (http://www.inf.puc-rio.br/~roberto/lpeg/re.html) +-- +-- @class module +-- @name re +-- @copyright 2008-2010 Lua.org, PUC-Rio. (https://svn.nmap.org/nmap/docs/licenses/MIT) + +--- Compiles the given string and returns an equivalent LPeg pattern. +-- +-- The given string may define either an expression or a grammar. The optional +-- defs table provides extra Lua values to be used by the pattern. +-- @class function +-- @name compile +-- @param string a regular expression or an LPeg grammar +-- @param defs Optional values to be used by the pattern +-- @return an LPeg pattern + +--- Searches the given pattern in the given subject. +-- +-- If it finds a match, returns the index where this occurrence starts and the +-- index where it ends. Otherwise, returns nil. +-- +-- An optional numeric argument init makes the search starts at that position +-- in the subject string. As usual in Lua libraries, a negative value counts +-- from the end. +-- @class function +-- @name find +-- @param subject The string to search +-- @param pattern A regular expression +-- @param init Optional index into subject to start searching +-- @return index where the occurrence starts or nil +-- @return index where the occurrence ends + +--- Global substitution. +-- +-- Does a global substitution, replacing all occurrences of pattern in the +-- given subject by replacement. +-- @class function +-- @name gsub +-- @param subject The string to search +-- @param pattern The pattern to match +-- @param replacement The replacement for each pattern +-- @return The string with all occurrences replaced + +--- Matches the given pattern against the given subject +-- +-- Matches the given pattern against the given subject, returning all captures. +-- @class function +-- @name match +-- @param subject The string to search +-- @param pattern The pattern to match +-- @param init Optional index into subject to start searching +-- @return pattern captures + +--- Updates the pre-defined character classes to the current locale. +-- +-- @class function +-- @name updatelocale + +-- $Id: re.lua,v 1.44 2013/03/26 20:11:40 roberto Exp $ + +-- imported functions and modules +local tonumber, type, print, error = tonumber, type, print, error +local setmetatable = setmetatable +local m = require"lpeg" + +-- 'm' will be used to parse expressions, and 'mm' will be used to +-- create expressions; that is, 're' runs on 'm', creating patterns +-- on 'mm' +local mm = m + +-- pattern's metatable +local mt = getmetatable(mm.P(0)) + + + +-- No more global accesses after this point +local version = _VERSION +if version == "Lua 5.2" then _ENV = nil end + + +local any = m.P(1) + + +-- Pre-defined names +local Predef = { nl = m.P"\n" } + + +local mem +local fmem +local gmem + + +local function updatelocale () + mm.locale(Predef) + Predef.a = Predef.alpha + Predef.c = Predef.cntrl + Predef.d = Predef.digit + Predef.g = Predef.graph + Predef.l = Predef.lower + Predef.p = Predef.punct + Predef.s = Predef.space + Predef.u = Predef.upper + Predef.w = Predef.alnum + Predef.x = Predef.xdigit + Predef.A = any - Predef.a + Predef.C = any - Predef.c + Predef.D = any - Predef.d + Predef.G = any - Predef.g + Predef.L = any - Predef.l + Predef.P = any - Predef.p + Predef.S = any - Predef.s + Predef.U = any - Predef.u + Predef.W = any - Predef.w + Predef.X = any - Predef.x + mem = {} -- restart memoization + fmem = {} + gmem = {} + local mt = {__mode = "v"} + setmetatable(mem, mt) + setmetatable(fmem, mt) + setmetatable(gmem, mt) +end + + +updatelocale() + + + +local I = m.P(function (s,i) print(i, s:sub(1, i-1)); return i end) + + +local function getdef (id, defs) + local c = defs and defs[id] + if not c then error("undefined name: " .. id) end + return c +end + + +local function patt_error (s, i) + local msg = (#s < i + 20) and s:sub(i) + or s:sub(i,i+20) .. "..." + msg = ("pattern error near '%s'"):format(msg) + error(msg, 2) +end + +local function mult (p, n) + local np = mm.P(true) + while n >= 1 do + if n%2 >= 1 then np = np * p end + p = p * p + n = n/2 + end + return np +end + +local function equalcap (s, i, c) + if type(c) ~= "string" then return nil end + local e = #c + i + if s:sub(i, e - 1) == c then return e else return nil end +end + + +local S = (Predef.space + "--" * (any - Predef.nl)^0)^0 + +local name = m.R("AZ", "az", "__") * m.R("AZ", "az", "__", "09")^0 + +local arrow = S * "<-" + +local seq_follow = m.P"/" + ")" + "}" + ":}" + "~}" + "|}" + (name * arrow) + -1 + +name = m.C(name) + + +-- a defined name only have meaning in a given environment +local Def = name * m.Carg(1) + +local num = m.C(m.R"09"^1) * S / tonumber + +local String = "'" * m.C((any - "'")^0) * "'" + + '"' * m.C((any - '"')^0) * '"' + + +local defined = "%" * Def / function (c,Defs) + local cat = Defs and Defs[c] or Predef[c] + if not cat then error ("name '" .. c .. "' undefined") end + return cat +end + +local Range = m.Cs(any * (m.P"-"/"") * (any - "]")) / mm.R + +local item = defined + Range + m.C(any) + +local Class = + "[" + * (m.C(m.P"^"^-1)) -- optional complement symbol + * m.Cf(item * (item - "]")^0, mt.__add) / + function (c, p) return c == "^" and any - p or p end + * "]" + +local function adddef (t, k, exp) + if t[k] then + error("'"..k.."' already defined as a rule") + else + t[k] = exp + end + return t +end + +local function firstdef (n, r) return adddef({n}, n, r) end + + +local function NT (n, b) + if not b then + error("rule '"..n.."' used outside a grammar") + else return mm.V(n) + end +end + + +local exp = m.P{ "Exp", + Exp = S * ( m.V"Grammar" + + m.Cf(m.V"Seq" * ("/" * S * m.V"Seq")^0, mt.__add) ); + Seq = m.Cf(m.Cc(m.P"") * m.V"Prefix"^0 , mt.__mul) + * (#seq_follow + patt_error); + Prefix = "&" * S * m.V"Prefix" / mt.__len + + "!" * S * m.V"Prefix" / mt.__unm + + m.V"Suffix"; + Suffix = m.Cf(m.V"Primary" * S * + ( ( m.P"+" * m.Cc(1, mt.__pow) + + m.P"*" * m.Cc(0, mt.__pow) + + m.P"?" * m.Cc(-1, mt.__pow) + + "^" * ( m.Cg(num * m.Cc(mult)) + + m.Cg(m.C(m.S"+-" * m.R"09"^1) * m.Cc(mt.__pow)) + ) + + "->" * S * ( m.Cg((String + num) * m.Cc(mt.__div)) + + m.P"{}" * m.Cc(nil, m.Ct) + + m.Cg(Def / getdef * m.Cc(mt.__div)) + ) + + "=>" * S * m.Cg(Def / getdef * m.Cc(m.Cmt)) + ) * S + )^0, function (a,b,f) return f(a,b) end ); + Primary = "(" * m.V"Exp" * ")" + + String / mm.P + + Class + + defined + + "{:" * (name * ":" + m.Cc(nil)) * m.V"Exp" * ":}" / + function (n, p) return mm.Cg(p, n) end + + "=" * name / function (n) return mm.Cmt(mm.Cb(n), equalcap) end + + m.P"{}" / mm.Cp + + "{~" * m.V"Exp" * "~}" / mm.Cs + + "{|" * m.V"Exp" * "|}" / mm.Ct + + "{" * m.V"Exp" * "}" / mm.C + + m.P"." * m.Cc(any) + + (name * -arrow + "<" * name * ">") * m.Cb("G") / NT; + Definition = name * arrow * m.V"Exp"; + Grammar = m.Cg(m.Cc(true), "G") * + m.Cf(m.V"Definition" / firstdef * m.Cg(m.V"Definition")^0, + adddef) / mm.P +} + +local pattern = S * m.Cg(m.Cc(false), "G") * exp / mm.P * (-any + patt_error) + + +local function compile (p, defs) + if mm.type(p) == "pattern" then return p end -- already compiled + local cp = pattern:match(p, 1, defs) + if not cp then error("incorrect pattern", 3) end + return cp +end + +local function match (s, p, i) + local cp = mem[p] + if not cp then + cp = compile(p) + mem[p] = cp + end + return cp:match(s, i or 1) +end + +local function find (s, p, i) + local cp = fmem[p] + if not cp then + cp = compile(p) / 0 + cp = mm.P{ mm.Cp() * cp * mm.Cp() + 1 * mm.V(1) } + fmem[p] = cp + end + local i, e = cp:match(s, i or 1) + if i then return i, e - 1 + else return i + end +end + +local function gsub (s, p, rep) + local g = gmem[p] or {} -- ensure gmem[p] is not collected while here + gmem[p] = g + local cp = g[rep] + if not cp then + cp = compile(p) + cp = mm.Cs((cp / rep + 1)^0) + g[rep] = cp + end + return cp:match(s) +end + + +-- exported names +local re = { + compile = compile, + match = match, + find = find, + gsub = gsub, + updatelocale = updatelocale, +} + +-- NSE uses Lua 5.2 +--if version == "Lua 5.1" then _G.re = re end + +return re -- cgit v1.2.3