-- -*- coding: raw-text-unix -*-
-- charset.lua - handle non-ascii characters and sgmlification.
-- This is part of blogme3.
-- Author: Eduardo Ochs <[email protected]>
-- Version: 2008jun30
-- <http://anggtwu.net/blogme3/charset.lua>
-- <http://anggtwu.net/blogme3/charset.lua.html>
-- License: GPL.
--
-- This is the part of blogme3 that handles weird (non-ascii)
-- characters. It works only for latin1 in unibyte encodings.
-- It supports just these glyphs by default: "*", "«", and "»".
-- (find-eevarticlesection "glyphs")
-- (find-blogme3 "anggdefs.lua" "eev_math_glyphs_edrx")

-- «.sgmlify»	(to "sgmlify")
-- «.unaccent»	(to "unaccent")


--------[ sgmlify ]--------
-- «sgmlify» (to ".sgmlify")
-- A test: lpeg.match(Sgmlifying, "Fun\231\227o")
-- (find-eoutput '(ascstr 128 255))
sgmlify_latin1_pairs = [[
  Æ AElig  Á Aacute  Acirc  À Agrave Å Aring  à Atilde Ä Auml  
  Ç Ccedil É Eacute Ê Ecirc  È Egrave Ë Euml   Í Iacute Ï Iuml  
  Ó Oacute Ô Ocirc  Ò Ograve Õ Otilde Ö Ouml   Ú Uacute Û Ucirc 
  Ù Ugrave Ü Uuml   á aacute â acirc  æ aelig  à agrave å aring 
  ã atilde ä auml   ç ccedil é eacute ê ecirc  è egrave ë euml  
  í iacute î icirc  ì igrave ï iuml   ó oacute ô ocirc  ò ograve
  õ otilde ö ouml   ß szlig  ú uacute û ucirc  ù ugrave ü uuml  
  ª ordf   « laquo  ° deg    º ordm   » raquo
  Ñ Ntilde ñ ntilde
  × times
]]
sgmlify_table = {}
for c,name in each2(split(sgmlify_latin1_pairs)) do
  sgmlify_table[c] = "&"..name..";"
end
sgmlify_table["\15"]  = "<font color=\"red\"><strong>*</strong></font>"
sgmlify_table["\171"] = "<font color=\"green\"><i>&laquo;</i></font>"
sgmlify_table["\187"] = "<font color=\"green\"><i>&raquo;</i></font>"
sgmlify_re            = "([\15\161-\254])"
sgmlify = function (s) return translatechars(s, sgmlify_re, sgmlify_table) end
-- (find-angg "LUA/lua50init.lua" "translatechars")


--------[ unaccent ]--------
-- «unaccent» (to ".unaccent")
-- (find-blogmefile "blogme2-outer.lua" "unaccent_from, unaccent_to =")

unaccent_from, unaccent_to = 
  "ÀÁÂÃÄÅÇÈÉÊË�Í�ÏÑÒÓÔÕÖ�ÙÚÛÜ�àáâãäåçèéêëìíîïñòóôõö�ùúûü�",
  "AAAAAACEEEEIIIINOOOOOOUUUUYaaaaaaceeeeiiiinoooooouuuuy"
unaccent_table = {}
for i = 1,strlen(unaccent_from) do
  unaccent_table[strsub(unaccent_from, i, i)] = strsub(unaccent_to, i, i)
end
unaccent_re = "([\192-\254])"
unaccent = function (str)
    return translatechars(str, unaccent_re, unaccent_table)
  end