iconvdata/ksc5601.h

   1 /* Access functions for KS C 5601-1992 based encoding conversion.
   2    Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4
   5    The GNU C Library is free software; you can redistribute it and/or
   6    modify it under the terms of the GNU Library General Public License as
   7    published by the Free Software Foundation; either version 2 of the
   8    License, or (at your option) any later version.
   9
  10    The GNU C Library is distributed in the hope that it will be useful,
  11    but WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13    Library General Public License for more details.
  14
  15    You should have received a copy of the GNU Library General Public
  16    License along with the GNU C Library; see the file COPYING.LIB.  If not,
  17    write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  18    Boston, MA 02111-1307, USA.  */
  19
  20 #ifndef _KSC5601_H
  21 #define _KSC5601_H      1
  22
  23 #define KSC5601_HANGUL 2350
  24 #define KSC5601_HANJA  4888
  25 #define KSC5601_SYMBOL  988
  26
  27 #include <gconv.h>
  28 #include <stdint.h>
  29
  30 /* Structure to map from UCS to KSC.  This structure should be packed
  31    on all platforms.  */
  32 struct map
  33 {
  34   uint16_t ucs;
  35   char val[2];
  36 };
  37
  38 /* Conversion table.  */
  39 extern const uint16_t __ksc5601_hangul_to_ucs[KSC5601_HANGUL];
  40 extern const uint16_t __ksc5601_sym_to_ucs[];
  41 extern const struct map __ksc5601_sym_from_ucs[KSC5601_SYMBOL];
  42 extern const uint16_t __ksc5601_hanja_to_ucs[KSC5601_HANJA];
  43 extern const struct map __ksc5601_hanja_from_ucs[KSC5601_HANJA];
  44
  45
  46 static inline uint32_t
  47 ksc5601_to_ucs4 (const unsigned char **s, size_t avail, unsigned char offset)
  48 {
  49   unsigned char ch = **s;
  50   unsigned char ch2;
  51   int idx;
  52
  53   /* row 94(0x7e) and row 41(0x49) are user-defined area in KS C 5601 */
  54
  55   if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) >= 0x7e
  56       || (ch - offset) == 0x49)
  57     return __UNKNOWN_10646_CHAR;
  58
  59   if (avail < 2)
  60     return 0;
  61
  62   ch2 = (*s)[1];
  63   if (ch2 < offset || (ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
  64     return __UNKNOWN_10646_CHAR;
  65
  66   idx = (ch - offset - 0x21) * 94 + (ch2 - offset - 0x21);
  67
  68   /* 1410 = 15 * 94 , 3760 = 40 * 94
  69      Hangul in KS C 5601 : row 16 - row 40 */
  70
  71   *s += 2;
  72
  73   if (idx >= 1410 && idx < 1410 + KSC5601_HANGUL)
  74     return (__ksc5601_hangul_to_ucs[idx - 1410]
  75             ?: (*s -= 2, __UNKNOWN_10646_CHAR));
  76   else if (idx >= 3854)
  77     /* Hanja : row 42 - row 93 : 3854 = 94 * (42-1) */
  78    return (__ksc5601_hanja_to_ucs[idx - 3854]
  79            ?: (*s -= 2, __UNKNOWN_10646_CHAR));
  80   else if (idx <= 1114)
  81     return __ksc5601_sym_to_ucs[idx] ?: (*s -= 2, __UNKNOWN_10646_CHAR);
  82
  83   *s -= 2;
  84   return __UNKNOWN_10646_CHAR;
  85 }
  86
  87 static inline size_t
  88 ucs4_to_ksc5601_hangul (uint32_t wch, unsigned char *s, size_t avail)
  89 {
  90   int l = 0;
  91   int u = KSC5601_HANGUL - 1;
  92   uint32_t try;
  93
  94   while (l <= u)
  95     {
  96       int m = (l + u) / 2;
  97       try = (uint32_t) __ksc5601_hangul_to_ucs[m];
  98       if (try > wch)
  99         u = m - 1;
 100       else if (try < wch)
 101         l= m + 1;
 102       else
 103         {
 104           if (avail < 2)
 105             return 0;
 106
 107           s[0] = (m / 94) + 0x30;
 108           s[1] = (m % 94) + 0x21;
 109
 110           return 2;
 111         }
 112     }
 113
 114   return __UNKNOWN_10646_CHAR;
 115 }
 116
 117
 118 static inline size_t
 119 ucs4_to_ksc5601_hanja (uint32_t wch, unsigned char *s, size_t avail)
 120 {
 121   int l = 0;
 122   int u = KSC5601_HANJA - 1;
 123   uint32_t try;
 124
 125   while (l <= u)
 126     {
 127       int m = (l + u) / 2;
 128       try = (uint32_t) __ksc5601_hanja_from_ucs[m].ucs;
 129       if (try > wch)
 130         u=m-1;
 131       else if (try < wch)
 132         l = m + 1;
 133       else
 134         {
 135           if (avail < 2)
 136             return 0;
 137
 138           s[0] = __ksc5601_hanja_from_ucs[m].val[0];
 139           s[1] = __ksc5601_hanja_from_ucs[m].val[1];
 140
 141           return 2;
 142         }
 143     }
 144
 145   return __UNKNOWN_10646_CHAR;
 146 }
 147
 148 static inline  size_t
 149 ucs4_to_ksc5601_sym (uint32_t wch, unsigned char *s, size_t avail)
 150 {
 151   int l = 0;
 152   int u = KSC5601_SYMBOL - 1;
 153   uint32_t try;
 154
 155   while (l <= u)
 156     {
 157       int m = (l + u) / 2;
 158       try = __ksc5601_sym_from_ucs[m].ucs;
 159       if (try > wch)
 160         u = m - 1;
 161       else if (try < wch)
 162         l = m + 1;
 163       else
 164         {
 165           if (avail < 2)
 166             return 0;
 167
 168           s[0] = __ksc5601_sym_from_ucs[m].val[0];
 169           s[1] = __ksc5601_sym_from_ucs[m].val[1];
 170
 171           return 2;
 172         }
 173     }
 174
 175   return __UNKNOWN_10646_CHAR;
 176 }
 177
 178
 179 static inline size_t
 180 ucs4_to_ksc5601 (uint32_t wch, unsigned char *s, size_t avail)
 181 {
 182   if (wch >= 0xac00 && wch <= 0xd7a3)
 183     return ucs4_to_ksc5601_hangul (wch, s, avail);
 184   else if ((wch >= 0x4e00 && wch <= 0x9fff)
 185            || (wch >= 0xf900 && wch <= 0xfa0b))
 186     return ucs4_to_ksc5601_hanja (wch, s, avail);
 187   else
 188     return ucs4_to_ksc5601_sym (wch, s, avail);
 189 }
 190
 191 #endif /* ksc5601.h */