=begin
XSD4R - Charset handling library.
Copyright (C) 2001, 2003  NAKAMURA, Hiroshi.

This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; either version 2 of the License, or (at your option) any later
version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY
WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
PRATICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc., 675 Mass
Ave, Cambridge, MA 02139, USA.
=end


module XSD


module Charset
  @encoding = $KCODE

  class XSDError < StandardError; end
  class CharsetError < XSDError; end
  class UnknownCharsetError < CharsetError; end
  class CharsetConversionError < CharsetError; end

public

  ###
  ## Maps
  #
  EncodingConvertMap = {}
  def Charset.init
    begin
      require 'xsd/iconvcharset'
      @encoding = 'UTF8'
      EncodingConvertMap[['UTF8', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "utf-8", str) }
      EncodingConvertMap[['EUC' , 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", "euc-jp", str) }
      EncodingConvertMap[['EUC' , 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv("shift-jis", "euc-jp", str) }
      if /(mswin|bccwin|mingw|cygwin)/ =~ RUBY_PLATFORM
	EncodingConvertMap[['UTF8', 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv("cp932", "utf-8", str) }
       	EncodingConvertMap[['SJIS', 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", "cp932", str) }
	EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "cp932", str) }
      else
	EncodingConvertMap[['UTF8', 'SJIS']] = Proc.new { |str| IconvCharset.safe_iconv("shift-jis", "utf-8", str) }
	EncodingConvertMap[['SJIS', 'UTF8']] = Proc.new { |str| IconvCharset.safe_iconv("utf-8", "shift-jis", str) }
	EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| IconvCharset.safe_iconv("euc-jp", "shift-jis", str) }
      end
    rescue LoadError
      begin
       	require 'nkf'
	EncodingConvertMap[['EUC' , 'SJIS']] = Proc.new { |str| NKF.nkf('-sXm0', str) }
	EncodingConvertMap[['SJIS', 'EUC' ]] = Proc.new { |str| NKF.nkf('-eXm0', str) }
      rescue LoadError
      end
  
      begin
	require 'uconv'
	@encoding = 'UTF8'
	EncodingConvertMap[['UTF8', 'EUC' ]] = Uconv.method(:u8toeuc)
	EncodingConvertMap[['UTF8', 'SJIS']] = Uconv.method(:u8tosjis)
	EncodingConvertMap[['EUC' , 'UTF8']] = Uconv.method(:euctou8)
	EncodingConvertMap[['SJIS', 'UTF8']] = Uconv.method(:sjistou8)
      rescue LoadError
      end
    end
  end
  self.init

  CharsetMap = {
    'NONE' => 'us-ascii',
    'EUC' => 'euc-jp',
    'SJIS' => 'shift_jis',
    'UTF8' => 'utf-8',
  }


  ###
  ## handlers
  #
  def Charset.encoding
    @encoding
  end

  def Charset.encoding_label
    charset_label(@encoding)
  end

  def Charset.encoding_to_xml(str, charset)
    encoding_conv(str, @encoding, charset_str(charset))
  end

  def Charset.encoding_from_xml(str, charset)
    encoding_conv(str, charset_str(charset), @encoding)
  end

  def Charset.encoding_conv(str, enc_from, enc_to)
    if enc_from == enc_to or enc_from == 'NONE' or enc_to == 'NONE'
      str
    elsif converter = EncodingConvertMap[[enc_from, enc_to]]
      converter.call(str)
    else
      raise CharsetConversionError.new(
	"Converter not found: #{ enc_from } -> #{ enc_to }")
    end
  end

  def Charset.charset_label(encoding)
    CharsetMap[encoding.upcase]
  end

  def Charset.charset_str(label)
    CharsetMap.index(label.downcase)
  end

  # us_ascii = '[\x00-\x7F]'
  us_ascii = '[\x9\xa\xd\x20-\x7F]'	# XML 1.0 restricted.
  USASCIIRegexp = Regexp.new("\\A#{ us_ascii }*\\z", nil, "NONE")

  twobytes_euc = '(?:[\x8E\xA1-\xFE][\xA1-\xFE])'
  threebytes_euc = '(?:\x8F[\xA1-\xFE][\xA1-\xFE])'
  character_euc = "(?:#{ us_ascii }|#{ twobytes_euc }|#{ threebytes_euc })"
  EUCRegexp = Regexp.new("\\A#{ character_euc }*\\z", nil, "NONE")

  # onebyte_sjis = '[\x00-\x7F\xA1-\xDF]'
  onebyte_sjis = '[\x9\xa\xd\x20-\x7F\xA1-\xDF]'	# XML 1.0 restricted.
  twobytes_sjis = '(?:[\x81-\x9F\xE0-\xFC][\x40-\x7E\x80-\xFC])'
  character_sjis = "(?:#{ onebyte_sjis }|#{ twobytes_sjis })"
  SJISRegexp = Regexp.new("\\A#{ character_sjis }*\\z", nil, "NONE")

  # 0xxxxxxx
  # 110yyyyy 10xxxxxx
  twobytes_utf8 = '(?:[\xC0-\xDF][\x80-\xBF])'
  # 1110zzzz 10yyyyyy 10xxxxxx
  threebytes_utf8 = '(?:[\xE0-\xEF][\x80-\xBF][\x80-\xBF])'
  # 11110uuu 10uuuzzz 10yyyyyy 10xxxxxx
  fourbytes_utf8 = '(?:[\xF0-\xF7][\x80-\xBF][\x80-\xBF][\x80-\xBF])'
  character_utf8 = "(?:#{ us_ascii }|#{ twobytes_utf8 }|#{ threebytes_utf8 }|#{ fourbytes_utf8 })"
  UTF8Regexp = Regexp.new("\\A#{ character_utf8 }*\\z", nil, "NONE")

  def Charset.is_us_ascii(str)
    USASCIIRegexp =~ str
  end

  def Charset.is_utf8(str)
    UTF8Regexp =~ str
  end

  def Charset.is_euc(str)
    EUCRegexp =~ str
  end

  def Charset.is_sjis(str)
    SJISRegexp =~ str
  end

  def Charset.is_ces(str, code = $KCODE)
    case code
    when 'NONE'
      is_us_ascii(str)
    when 'UTF8'
      is_utf8(str)
    when 'EUC'
      is_euc(str)
    when 'SJIS'
      is_sjis(str)
    else
      raise UnknownCharsetError.new("Unknown charset: #{ code }")
    end
  end
end


end