iconvdata/ksc5601.h

   1 /* Access functions for KS C 5601-1992 based encoding conversion.
   2    Copyright (C) 1998, 1999, 2000, 2003 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 Lesser General Public
   7    License as published by the Free Software Foundation; either
   8    version 2.1 of the 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    Lesser General Public License for more details.
  14
  15    You should have received a copy of the GNU Lesser General Public
  16    License along with the GNU C Library; if not, write to the Free
  17    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  18    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 __attribute ((always_inline))
  48 ksc5601_to_ucs4 (const unsigned char **s, size_t avail, unsigned char offset)
  49 {
  50   unsigned char ch = **s;
  51   unsigned char ch2;
  52   int idx;
  53
  54   /* row 94(0x7e) and row 41(0x49) are user-defined area in KS C 5601 */
  55
  56   if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) >= 0x7e
  57       || (ch - offset) == 0x49)
  58     return __UNKNOWN_10646_CHAR;
  59
  60   if (avail < 2)
  61     return 0;
  62
  63   ch2 = (*s)[1];
  64   if (ch2 < offset || (ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
  65     return __UNKNOWN_10646_CHAR;
  66
  67   idx = (ch - offset - 0x21) * 94 + (ch2 - offset - 0x21);
  68
  69   /* 1410 = 15 * 94 , 3760 = 40 * 94
  70      Hangul in KS C 5601 : row 16 - row 40 */
  71
  72   *s += 2;
  73
  74   if (idx >= 1410 && idx < 1410 + KSC5601_HANGUL)
  75     return (__ksc5601_hangul_to_ucs[idx - 1410]
  76             ?: (*s -= 2, __UNKNOWN_10646_CHAR));
  77   else if (idx >= 3854)
  78     /* Hanja : row 42 - row 93 : 3854 = 94 * (42-1) */
  79    return (__ksc5601_hanja_to_ucs[idx - 3854]
  80            ?: (*s -= 2, __UNKNOWN_10646_CHAR));
  81   else if (idx <= 1114)
  82     return __ksc5601_sym_to_ucs[idx] ?: (*s -= 2, __UNKNOWN_10646_CHAR);
  83
  84   *s -= 2;
  85   return __UNKNOWN_10646_CHAR;
  86 }
  87
  88 static inline size_t
  89 __attribute ((always_inline))
  90 ucs4_to_ksc5601_hangul (uint32_t wch, unsigned char *s, size_t avail)
  91 {
  92   int l = 0;
  93   int u = KSC5601_HANGUL - 1;
  94   uint32_t try;
  95
  96   while (l <= u)
  97     {
  98       int m = (l + u) / 2;
  99       try = (uint32_t) __ksc5601_hangul_to_ucs[m];
 100       if (try > wch)
 101         u = m - 1;
 102       else if (try < wch)
 103         l= m + 1;
 104       else
 105         {
 106           if (avail < 2)
 107             return 0;
 108
 109           s[0] = (m / 94) + 0x30;
 110           s[1] = (m % 94) + 0x21;
 111
 112           return 2;
 113         }
 114     }
 115
 116   return __UNKNOWN_10646_CHAR;
 117 }
 118
 119
 120 static inline size_t
 121 __attribute ((always_inline))
 122 ucs4_to_ksc5601_hanja (uint32_t wch, unsigned char *s, size_t avail)
 123 {
 124   int l = 0;
 125   int u = KSC5601_HANJA - 1;
 126   uint32_t try;
 127
 128   while (l <= u)
 129     {
 130       int m = (l + u) / 2;
 131       try = (uint32_t) __ksc5601_hanja_from_ucs[m].ucs;
 132       if (try > wch)
 133         u=m-1;
 134       else if (try < wch)
 135         l = m + 1;
 136       else
 137         {
 138           if (avail < 2)
 139             return 0;
 140
 141           s[0] = __ksc5601_hanja_from_ucs[m].val[0];
 142           s[1] = __ksc5601_hanja_from_ucs[m].val[1];
 143
 144           return 2;
 145         }
 146     }
 147
 148   return __UNKNOWN_10646_CHAR;
 149 }
 150
 151 static inline  size_t
 152 __attribute ((always_inline))
 153 ucs4_to_ksc5601_sym (uint32_t wch, unsigned char *s, size_t avail)
 154 {
 155   int l = 0;
 156   int u = KSC5601_SYMBOL - 1;
 157   uint32_t try;
 158
 159   while (l <= u)
 160     {
 161       int m = (l + u) / 2;
 162       try = __ksc5601_sym_from_ucs[m].ucs;
 163       if (try > wch)
 164         u = m - 1;
 165       else if (try < wch)
 166         l = m + 1;
 167       else
 168         {
 169           if (avail < 2)
 170             return 0;
 171
 172           s[0] = __ksc5601_sym_from_ucs[m].val[0];
 173           s[1] = __ksc5601_sym_from_ucs[m].val[1];
 174
 175           return 2;
 176         }
 177     }
 178
 179   return __UNKNOWN_10646_CHAR;
 180 }
 181
 182
 183 static inline size_t
 184 __attribute ((always_inline))
 185 ucs4_to_ksc5601 (uint32_t wch, unsigned char *s, size_t avail)
 186 {
 187   if (wch >= 0xac00 && wch <= 0xd7a3)
 188     return ucs4_to_ksc5601_hangul (wch, s, avail);
 189   else if ((wch >= 0x4e00 && wch <= 0x9fff)
 190            || (wch >= 0xf900 && wch <= 0xfa0b))
 191     return ucs4_to_ksc5601_hanja (wch, s, avail);
 192   else
 193     return ucs4_to_ksc5601_sym (wch, s, avail);
 194 }
 195
 196 #endif /* ksc5601.h */