summaryrefslogtreecommitdiff
path: root/Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua
diff options
context:
space:
mode:
authorKarl Berry <karl@freefriends.org>2013-05-05 22:52:10 +0000
committerKarl Berry <karl@freefriends.org>2013-05-05 22:52:10 +0000
commit5f9dd2add16775cc6aece96e2695c7e475ee1834 (patch)
treefba8f32362b01a41db0125ab430fd1962b290b4f /Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua
parentc275a5f0b045f10e93c6a557f2499f3cf7709efc (diff)
lualibs (5may13)
git-svn-id: svn://tug.org/texlive/trunk@30238 c570f23f-e606-0410-a88d-b1316a301751
Diffstat (limited to 'Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua')
-rw-r--r--Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua399
1 files changed, 158 insertions, 241 deletions
diff --git a/Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua b/Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua
index 9856d521249..77c076cc53e 100644
--- a/Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua
+++ b/Master/texmf-dist/tex/luatex/lualibs/lualibs-string.lua
@@ -6,283 +6,200 @@ if not modules then modules = { } end modules ['l-string'] = {
license = "see context related readme files"
}
-local sub, gsub, find, match, gmatch, format, char, byte, rep, lower = string.sub, string.gsub, string.find, string.match, string.gmatch, string.format, string.char, string.byte, string.rep, string.lower
-local lpegmatch = lpeg.match
-
--- some functions may disappear as they are not used anywhere
-
-if not string.split then
-
- -- this will be overloaded by a faster lpeg variant
-
- function string:split(pattern)
- if #self > 0 then
- local t = { }
- for s in gmatch(self..pattern,"(.-)"..pattern) do
- t[#t+1] = s
- end
- return t
- else
- return { }
- end
- end
-
-end
-
-local chr_to_esc = {
- ["%"] = "%%",
- ["."] = "%.",
- ["+"] = "%+", ["-"] = "%-", ["*"] = "%*",
- ["^"] = "%^", ["$"] = "%$",
- ["["] = "%[", ["]"] = "%]",
- ["("] = "%(", [")"] = "%)",
- ["{"] = "%{", ["}"] = "%}"
-}
-
-string.chr_to_esc = chr_to_esc
-
-function string:esc() -- variant 2
- return (gsub(self,"(.)",chr_to_esc))
-end
-
-function string:unquote()
- return (gsub(self,"^([\"\'])(.*)%1$","%2"))
-end
-
---~ function string:unquote()
---~ if find(self,"^[\'\"]") then
---~ return sub(self,2,-2)
---~ else
---~ return self
---~ end
---~ end
-
-function string:quote() -- we could use format("%q")
- return format("%q",self)
-end
-
-function string:count(pattern) -- variant 3
+local string = string
+local sub, gmatch, format, char, byte, rep, lower = string.sub, string.gmatch, string.format, string.char, string.byte, string.rep, string.lower
+local lpegmatch, patterns = lpeg.match, lpeg.patterns
+local P, S, C, Ct, Cc, Cs = lpeg.P, lpeg.S, lpeg.C, lpeg.Ct, lpeg.Cc, lpeg.Cs
+
+-- Some functions are already defined in l-lpeg and maybe some from here will
+-- move there (unless we also expose caches).
+
+-- if not string.split then
+--
+-- function string.split(str,pattern)
+-- local t = { }
+-- if #str > 0 then
+-- local n = 1
+-- for s in gmatch(str..pattern,"(.-)"..pattern) do
+-- t[n] = s
+-- n = n + 1
+-- end
+-- end
+-- return t
+-- end
+--
+-- end
+
+-- function string.unquoted(str)
+-- return (gsub(str,"^([\"\'])(.*)%1$","%2")) -- interesting pattern
+-- end
+
+local unquoted = patterns.squote * C(patterns.nosquote) * patterns.squote
+ + patterns.dquote * C(patterns.nodquote) * patterns.dquote
+
+function string.unquoted(str)
+ return lpegmatch(unquoted,str) or str
+end
+
+-- print(string.unquoted("test"))
+-- print(string.unquoted([["t\"est"]]))
+-- print(string.unquoted([["t\"est"x]]))
+-- print(string.unquoted("\'test\'"))
+-- print(string.unquoted('"test"'))
+-- print(string.unquoted('"test"'))
+
+function string.quoted(str)
+ return format("%q",str) -- always double quote
+end
+
+function string.count(str,pattern) -- variant 3
local n = 0
- for _ in gmatch(self,pattern) do
+ for _ in gmatch(str,pattern) do -- not for utf
n = n + 1
end
return n
end
-function string:limit(n,sentinel)
- if #self > n then
- sentinel = sentinel or " ..."
- return sub(self,1,(n-#sentinel)) .. sentinel
+function string.limit(str,n,sentinel) -- not utf proof
+ if #str > n then
+ sentinel = sentinel or "..."
+ return sub(str,1,(n-#sentinel)) .. sentinel
else
- return self
+ return str
end
end
---~ function string:strip() -- the .- is quite efficient
---~ -- return match(self,"^%s*(.-)%s*$") or ""
---~ -- return match(self,'^%s*(.*%S)') or '' -- posted on lua list
---~ return find(s,'^%s*$') and '' or match(s,'^%s*(.*%S)')
---~ end
-
-do -- roberto's variant:
- local space = lpeg.S(" \t\v\n")
- local nospace = 1 - space
- local stripper = space^0 * lpeg.C((space^0 * nospace^1)^0)
- function string.strip(str)
- return lpegmatch(stripper,str) or ""
- end
-end
-
-function string:is_empty()
- return not find(self,"%S")
-end
+local stripper = patterns.stripper
+local collapser = patterns.collapser
+local longtostring = patterns.longtostring
-function string:enhance(pattern,action)
- local ok, n = true, 0
- while ok do
- ok = false
- self = gsub(self,pattern, function(...)
- ok, n = true, n + 1
- return action(...)
- end)
- end
- return self, n
+function string.strip(str)
+ return lpegmatch(stripper,str) or ""
end
-local chr_to_hex, hex_to_chr = { }, { }
-
-for i=0,255 do
- local c, h = char(i), format("%02X",i)
- chr_to_hex[c], hex_to_chr[h] = h, c
+function string.collapsespaces(str)
+ return lpegmatch(collapser,str) or ""
end
-function string:to_hex()
- return (gsub(self or "","(.)",chr_to_hex))
+function string.longtostring(str)
+ return lpegmatch(longtostring,str) or ""
end
-function string:from_hex()
- return (gsub(self or "","(..)",hex_to_chr))
-end
+-- function string.is_empty(str)
+-- return not find(str,"%S")
+-- end
-if not string.characters then
+local pattern = P(" ")^0 * P(-1)
- local function nextchar(str, index)
- index = index + 1
- return (index <= #str) and index or nil, sub(str,index,index)
- end
- function string:characters()
- return nextchar, self, 0
- end
- local function nextbyte(str, index)
- index = index + 1
- return (index <= #str) and index or nil, byte(sub(str,index,index))
- end
- function string:bytes()
- return nextbyte, self, 0
- end
-
-end
-
--- we can use format for this (neg n)
-
-function string:rpadd(n,chr)
- local m = n-#self
- if m > 0 then
- return self .. rep(chr or " ",m)
+function string.is_empty(str)
+ if str == "" then
+ return true
else
- return self
- end
-end
-
-function string:lpadd(n,chr)
- local m = n-#self
- if m > 0 then
- return rep(chr or " ",m) .. self
+ return lpegmatch(pattern,str) and true or false
+ end
+end
+
+-- if not string.escapedpattern then
+--
+-- local patterns_escapes = {
+-- ["%"] = "%%",
+-- ["."] = "%.",
+-- ["+"] = "%+", ["-"] = "%-", ["*"] = "%*",
+-- ["["] = "%[", ["]"] = "%]",
+-- ["("] = "%(", [")"] = "%)",
+-- -- ["{"] = "%{", ["}"] = "%}"
+-- -- ["^"] = "%^", ["$"] = "%$",
+-- }
+--
+-- local simple_escapes = {
+-- ["-"] = "%-",
+-- ["."] = "%.",
+-- ["?"] = ".",
+-- ["*"] = ".*",
+-- }
+--
+-- function string.escapedpattern(str,simple)
+-- return (gsub(str,".",simple and simple_escapes or patterns_escapes))
+-- end
+--
+-- function string.topattern(str,lowercase,strict)
+-- if str == "" then
+-- return ".*"
+-- else
+-- str = gsub(str,".",simple_escapes)
+-- if lowercase then
+-- str = lower(str)
+-- end
+-- if strict then
+-- return "^" .. str .. "$"
+-- else
+-- return str
+-- end
+-- end
+-- end
+--
+-- end
+
+--- needs checking
+
+local anything = patterns.anything
+local allescapes = Cc("%") * S(".-+%?()[]*") -- also {} and ^$ ?
+local someescapes = Cc("%") * S(".-+%()[]") -- also {} and ^$ ?
+local matchescapes = Cc(".") * S("*?") -- wildcard and single match
+
+local pattern_a = Cs ( ( allescapes + anything )^0 )
+local pattern_b = Cs ( ( someescapes + matchescapes + anything )^0 )
+local pattern_c = Cs ( Cc("^") * ( someescapes + matchescapes + anything )^0 * Cc("$") )
+
+function string.escapedpattern(str,simple)
+ return lpegmatch(simple and pattern_b or pattern_a,str)
+end
+
+function string.topattern(str,lowercase,strict)
+ if str=="" or type(str) ~= "string" then
+ return ".*"
+ elseif strict then
+ str = lpegmatch(pattern_c,str)
else
- return self
+ str = lpegmatch(pattern_b,str)
end
-end
-
-string.padd = string.rpadd
-
-function is_number(str) -- tonumber
- return find(str,"^[%-%+]?[%d]-%.?[%d+]$") == 1
-end
-
---~ print(is_number("1"))
---~ print(is_number("1.1"))
---~ print(is_number(".1"))
---~ print(is_number("-0.1"))
---~ print(is_number("+0.1"))
---~ print(is_number("-.1"))
---~ print(is_number("+.1"))
-
-function string:split_settings() -- no {} handling, see l-aux for lpeg variant
- if find(self,"=") then
- local t = { }
- for k,v in gmatch(self,"(%a+)=([^%,]*)") do
- t[k] = v
- end
- return t
+ if lowercase then
+ return lower(str)
else
- return nil
+ return str
end
end
-local patterns_escapes = {
- ["-"] = "%-",
- ["."] = "%.",
- ["+"] = "%+",
- ["*"] = "%*",
- ["%"] = "%%",
- ["("] = "%)",
- [")"] = "%)",
- ["["] = "%[",
- ["]"] = "%]",
-}
+-- print(string.escapedpattern("12+34*.tex",false))
+-- print(string.escapedpattern("12+34*.tex",true))
+-- print(string.topattern ("12+34*.tex",false,false))
+-- print(string.topattern ("12+34*.tex",false,true))
-function string:pattesc()
- return (gsub(self,".",patterns_escapes))
+function string.valid(str,default)
+ return (type(str) == "string" and str ~= "" and str) or default or nil
end
-local simple_escapes = {
- ["-"] = "%-",
- ["."] = "%.",
- ["?"] = ".",
- ["*"] = ".*",
-}
+-- handy fallback
-function string:simpleesc()
- return (gsub(self,".",simple_escapes))
-end
+string.itself = function(s) return s end
-function string:tohash()
- local t = { }
- for s in gmatch(self,"([^, ]+)") do -- lpeg
- t[s] = true
- end
- return t
-end
+-- also handy (see utf variant)
-local pattern = lpeg.Ct(lpeg.C(1)^0)
+local pattern = Ct(C(1)^0) -- string and not utf !
-function string:totable()
- return lpegmatch(pattern,self)
+function string.totable(str)
+ return lpegmatch(pattern,str)
end
---~ local t = {
---~ "1234567123456712345671234567",
---~ "a\tb\tc",
---~ "aa\tbb\tcc",
---~ "aaa\tbbb\tccc",
---~ "aaaa\tbbbb\tcccc",
---~ "aaaaa\tbbbbb\tccccc",
---~ "aaaaaa\tbbbbbb\tcccccc",
---~ }
---~ for k,v do
---~ print(string.tabtospace(t[k]))
---~ end
-
-function string.tabtospace(str,tab)
- -- we don't handle embedded newlines
- while true do
- local s = find(str,"\t")
- if s then
- if not tab then tab = 7 end -- only when found
- local d = tab-(s-1) % tab
- if d > 0 then
- str = gsub(str,"\t",rep(" ",d),1)
- else
- str = gsub(str,"\t","",1)
- end
- else
- break
- end
- end
- return str
-end
+-- handy from within tex:
-function string:compactlong() -- strips newlines and leading spaces
- self = gsub(self,"[\n\r]+ *","")
- self = gsub(self,"^ *","")
- return self
-end
+local replacer = lpeg.replacer("@","%%") -- Watch the escaped % in lpeg!
-function string:striplong() -- strips newlines and leading spaces
- self = gsub(self,"^%s*","")
- self = gsub(self,"[\n\r]+ *","\n")
- return self
+function string.tformat(fmt,...)
+ return format(lpegmatch(replacer,fmt),...)
end
-function string:topattern(lowercase,strict)
- if lowercase then
- self = lower(self)
- end
- self = gsub(self,".",simple_escapes)
- if self == "" then
- self = ".*"
- elseif strict then
- self = "^" .. self .. "$"
- end
- return self
-end
+-- obsolete names:
+
+string.quote = string.quoted
+string.unquote = string.unquoted