iconvdata/ksc5601.h

   1 /* Access functions for KS C 5601-1992 based encoding conversion.
   2    Copyright (C) 1998, 1999 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  986
  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   return __UNKNOWN_10646_CHAR;
  84 }
  85
  86 static inline size_t
  87 ucs4_to_ksc5601_hangul (uint32_t wch, unsigned char *s, size_t avail)
  88 {
  89   int l = 0;
  90   int u = KSC5601_HANGUL - 1;
  91   uint32_t try;
  92
  93   while (l <= u)
  94     {
  95       int m = (l + u) / 2;
  96       try = (uint32_t) __ksc5601_hangul_to_ucs[m];
  97       if (try > wch)
  98         u = m - 1;
  99       else if (try < wch)
 100         l= m + 1;
 101       else
 102         {
 103           if (avail < 2)
 104             return 0;
 105
 106           s[0] = (m / 94) + 0x30;
 107           s[1] = (m % 94) + 0x21;
 108
 109           return 2;
 110         }
 111     }
 112
 113   return __UNKNOWN_10646_CHAR;
 114 }
 115
 116
 117 static inline size_t
 118 ucs4_to_ksc5601_hanja (uint32_t wch, unsigned char *s, size_t avail)
 119 {
 120   int l = 0;
 121   int u = KSC5601_HANJA - 1;
 122   uint32_t try;
 123
 124   while (l <= u)
 125     {
 126       int m = (l + u) / 2;
 127       try = (uint32_t) __ksc5601_hanja_from_ucs[m].ucs;
 128       if (try > wch)
 129         u=m-1;
 130       else if (try < wch)
 131         l = m + 1;
 132       else
 133         {
 134           if (avail < 2)
 135             return 0;
 136
 137           s[0] = __ksc5601_hanja_from_ucs[m].val[0];
 138           s[1] = __ksc5601_hanja_from_ucs[m].val[1];
 139
 140           return 2;
 141         }
 142     }
 143
 144   return __UNKNOWN_10646_CHAR;
 145 }
 146
 147 static inline  size_t
 148 ucs4_to_ksc5601_sym (uint32_t wch, unsigned char *s, size_t avail)
 149 {
 150   int l = 0;
 151   int u = KSC5601_SYMBOL - 1;
 152   uint32_t try;
 153
 154   while (l <= u)
 155     {
 156       int m = (l + u) / 2;
 157       try = __ksc5601_sym_from_ucs[m].ucs;
 158       if (try > wch)
 159         u = m - 1;
 160       else if (try < wch)
 161         l = m + 1;
 162       else
 163         {
 164           if (avail < 2)
 165             return 0;
 166
 167           s[0] = __ksc5601_sym_from_ucs[m].val[0];
 168           s[1] = __ksc5601_sym_from_ucs[m].val[1];
 169
 170           return 2;
 171         }
 172     }
 173
 174   return __UNKNOWN_10646_CHAR;
 175 }
 176
 177
 178 static inline size_t
 179 ucs4_to_ksc5601 (uint32_t wch, unsigned char *s, size_t avail)
 180 {
 181   if (wch >= 0xac00 && wch <= 0xd7a3)
 182     return ucs4_to_ksc5601_hangul (wch, s, avail);
 183   else if ((wch >= 0x4e00 && wch <= 0x9fff)
 184            || (wch >= 0xf900 && wch <= 0xfa0b))
 185     return ucs4_to_ksc5601_hanja (wch, s, avail);
 186   else
 187     return ucs4_to_ksc5601_sym (wch, s, avail);
 188 }
 189
 190 #endif /* ksc5601.h */