kanjiinfo.h

Displays the Japanese text

#ifndef _KANJIINFO_H_
#define _KANJIINFO_H_

#include <iostream>

/**
 * @class KanjiInfo
 *
 */
class KanjiInfo {

 public:

  typedef unsigned short KANJI;

  enum ENCODING { JIS, SJIS, EUC };

	static unsigned short countKanjiUnicode();
	static unsigned short countMiscUnicode();
	static unsigned short countExtUnicode();


 public:

	unsigned short jis2sjis (unsigned short);
	unsigned short jis2unicode (unsigned short);
	unsigned short sjis2jis (unsigned short);
	unsigned short unicode2jis (unsigned short);

 private:

	void sjis2jis (unsigned short *,unsigned short *);

  // Code pages for various locales
  static KANJI misc_unicode[];
  static KANJI ext_unicode[];
  static KANJI kanji_unicode[];

  static KANJI cp1250[];
  static KANJI cp1251[];
  static KANJI cp1252[];
  static KANJI cp1253[];
  static KANJI cp1254[];
  static KANJI cp1255[];
  static KANJI cp1256[];
  static KANJI cp1257[];
  static KANJI cp1258[];  

};

/**
 * @class jis
 * 
 * STL iostream manipulator to convert character given in JIS 
 * encoding to EUC or SJIS.
 *
 * Greatly simplifies the conversion process and makes it a more
 * locigal implementation.
 *
 * Example usage:
 *   cout << jis_code << ",  EUC: " << jis<EUC>(jis_code) << endl;
 *   cout << jis_code << ", SJIS: " << jis<SJIS>(jis_code) << endl;
 */
template <KanjiInfo::ENCODING enc = EUC>
class jis : KanjiInfo {
  
  public:
  explicit jis(unsigned short jis_code) {
    
    switch(enc) {
      case KanjiInfo::EUC:
        _code = convert_unicode(jis_code);
        break;
      case KanjiInfo::SJIS:
        _code = convert_sjis(jis_code);
        break;
      default:
      _code = 0;
    }
    
  }
  
  private:
  
  unsigned short _code;
  
  /**
   * Convert JIS to Unicode
   *
   * Based on iformation given in [1, p.262]
   *
   * 0x0021->0x007E  : ASCII
   * 0x00A1->0x00DF  : 1/2 width Katakana
   *
   * 0x0081->0x009F  : 
   * 0x00E0->0x00EF  : @see JIS x 0208-1990
   * 0x0040->0x007E
   * 0x0080->0x00FC  : @see JIS x 0208-1990
   */
  unsigned short convert_unicode (unsigned short ch) {
    
    if(ch <= 0x7e)
      return (ch); // ASCII
    
    if((ch >= 0x0080) && (ch <= 0x00ff))
      return (ext_unicode[ch-0x80]);  // Extended ASCII
    
    if((ch >= 0x2330) && (ch <= 0x237a)) 
      return (ch-0x2330+0xff10);      // Japanese ASCII
    
    if((ch >= 0x2421) && (ch <= 0x2473)) 
      return (ch-0x2421+0x3041);      // Hiragana
    
    if((ch >= 0x2521) && (ch <= 0x2576)) 
      return (ch-0x2521+0x30a1);      // Katakana
    
    if((ch >= 0x2621) && (ch <= 0x2658)) 
      return (ch-0x2621+0x0391);      // Greek
    
    if((ch >= 0x2721) && (ch <= 0x2771)) 
      return (ch-0x2721+0x0410);      
    
    // Cyrillic
    if((ch >= 0x3021) && (ch <= 0x7426)) {
      // Kanji    
      ch -= 0x3021;
      ch  = ((ch&0xff00)>>0x08) * (94+(ch&0x00ff));
      return (kanji_unicode[ch]);
    }
    
    if((ch >= 0x2121) && (ch <= 0x217e)) 
      return (misc_unicode[ch-0x2121]);       // Main symbol group
    
    if ((ch >= 0x2221) && (ch <= 0x227f)) 
      return (misc_unicode[ch-0x2221+94]);    // Secondary symbol group
    
    return (0);
  }
  
  /**
   * Convert JIS to SJIS
   *
   */
  unsigned short convert_sjis(unsigned short ch) {
    
    unsigned short hi, lo;

    hi = (ch >> 8);
    lo = (ch & 0xff);

    hi =  ((hi - ((hi < 160) ? 112 : 176)) << 1);
    if(lo < 159) hi--;
    lo -= ( (lo < 159) ? ((lo > 127) ? 32 : 31) : 126);
    
    return ((hi << 8) | lo);
  }

  /**
   * Perform the output manipultion
   */
  template <class charT, class Traits>
    inline friend std::basic_ostream<charT,Traits>& operator<<(std::basic_ostream<charT,Traits>& os, const jis& j) {

    if(_code)
      os.put(_code);
  }

};


/**
 * @class euc
 * 
 * STL iostream manipulator to convert character given in EUC 
 * encoding to JIS or SJIS.
 *
 * Greatly simplifies the conversion process and makes it a more
 * locigal implementation.
 *
 */
template <KanjiInfo::ENCODING enc = EUC>
class euc : KanjiInfo {
  
  public:
  explicit euc(unsigned short euc_code) {
    
    switch(enc) {
      case KanjiInfo::JIS:
        _code = convert_jis(euc_code);
        break;
      case KanjiInfo::SJIS:
        _code = convert_sjis(euc_code);
        break;
      default:
      _code = 0;
    }
    
  }
  
  private:
  
  unsigned short _code;

  unsigned short convert_jis(unsigned short ch) {

		unsigned short i;

		if((ch <= 0x007e)) 
			return (ch);                // ASCII

		if((ch >= 0x3041) && (ch <= 0x3093)) 
			return (ch-0x3041+0x2421);  // Hiragana

		if((ch >= 0x30a1) && (ch <= 0x30f6)) 
			return (ch-0x30a1+0x2521);  // Katakana

		if((ch >= 0x0391) && (ch <= 0x03c9)) 
			return (ch-0x0391+0x2621);  // Greek

		if((ch >= 0x0410) && (ch <= 0x044f))
			return (ch-0x0410+0x2721);  // Cyrillic

		// The kanji
		for(i=0; i < countKanjiUnicode(); i++) {                       

			if (kanji_unicode[i] == ch) {
		    ch = i/94;
			  return (((ch+0x30) << 8) | (i-ch*94)+0x21);
			}

		}

		// Symbol and pucntuation (main group)
		for (i=0; i< countMiscUnicode(); i++) {                        

			if (misc_unicode[i] == ch) return ((i >= 94) ? 0x2221+i-94 : 0x2121+i);

		}

		// Extneded ascii codes.
		for (i=0; i<NUMBER_EXTUNICODE; i++) {
			if (ext_unicode[i] == ch) 
				return (i+0x80);
		}
	
		if((ch >= 0xff10) && (ch <= 0xff5a)) 
			return (ch-0xff10+0x2330);  // Japanese ASCII.
		
		// The Japanese ASCII needs to be here because some puncutation is stored within gaps
		return 0;

	}
  
  unsigned short convert_sjis(unsigned short ch) {
	}
	
	/**
   * Perform the output manipultion
   */
  template <class charT, class Traits>
    inline friend std::basic_ostream<charT,Traits>& operator<<(std::basic_ostream<charT,Traits>& os, const jis& j) {

    if(_code)
      os.put(_code);
  }

};

#endif