iconvdata/cns11643.h

   1 /* Access functions for CNS 11643, plane 2 handling.
   2    Copyright (C) 1998, 1999 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4    Contributed by Ulrich Drepper <drepper@cygnus.com>, 1998.
   5
   6    The GNU C Library is free software; you can redistribute it and/or
   7    modify it under the terms of the GNU Library General Public License as
   8    published by the Free Software Foundation; either version 2 of the
   9    License, or (at your option) any later version.
  10
  11    The GNU C Library is distributed in the hope that it will be useful,
  12    but WITHOUT ANY WARRANTY; without even the implied warranty of
  13    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14    Library General Public License for more details.
  15
  16    You should have received a copy of the GNU Library General Public
  17    License along with the GNU C Library; see the file COPYING.LIB.  If not,
  18    write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  19    Boston, MA 02111-1307, USA.  */
  20
  21 #include <stdint.h>
  22
  23 /* Table for CNS 11643, plane 2 to UCS4 conversion.  */
  24 extern const uint16_t __cns11643l2_to_ucs4_tab[];
  25 extern const uint16_t __cns11643l14_to_ucs4_tab[];
  26
  27
  28 static inline uint32_t
  29 cns11643_to_ucs4 (const char **s, size_t avail, unsigned char offset)
  30 {
  31   unsigned char ch = *(*s);
  32   unsigned char ch2;
  33   unsigned char ch3;
  34   uint32_t result;
  35   int idx;
  36
  37   if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) > 0x30)
  38     return __UNKNOWN_10646_CHAR;
  39
  40   if (avail < 3)
  41     return 0;
  42
  43   ch2 = (*s)[1];
  44   if ((ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
  45     return __UNKNOWN_10646_CHAR;
  46
  47   ch3 = (*s)[2];
  48   if ((ch3 - offset) <= 0x20 || (ch3 - offset) >= 0x7f)
  49     return __UNKNOWN_10646_CHAR;
  50
  51   idx = (ch2 - 0x21 - offset) * 94 + (ch3 - 0x21 - offset);
  52
  53   if ((ch - 0x21 - offset) == 1)
  54     {
  55       if (idx > 0x2196)
  56         return __UNKNOWN_10646_CHAR;
  57       result = __cns11643l1_to_ucs4_tab[idx];
  58     }
  59   else if ((ch - 0x21 - offset) == 2)
  60     {
  61       if (idx > 0x1de1)
  62         return __UNKNOWN_10646_CHAR;
  63       result = __cns11643l2_to_ucs4_tab[idx];
  64     }
  65   else if ((ch - 0x21 - offset) == 0xe)
  66     {
  67       if (idx > 0x19bd)
  68         return __UNKNOWN_10646_CHAR;
  69       result = __cns11643l14_to_ucs4_tab[idx];
  70     }
  71   else
  72     return __UNKNOWN_10646_CHAR;
  73
  74   if (result != L'\0')
  75     (*s) += 3;
  76   else
  77     result = __UNKNOWN_10646_CHAR;
  78
  79   return result;
  80 }
  81
  82
  83 /* Tables for the UCS4 -> CNS conversion.  */
  84 extern const char __cns11643l1_from_ucs4_tab1[][2];
  85 extern const char __cns11643l1_from_ucs4_tab2[][2];
  86 extern const char __cns11643l1_from_ucs4_tab3[][2];
  87 extern const char __cns11643l1_from_ucs4_tab4[][2];
  88 extern const char __cns11643l1_from_ucs4_tab5[][2];
  89 extern const char __cns11643l1_from_ucs4_tab6[][2];
  90 extern const char __cns11643l1_from_ucs4_tab7[][2];
  91 extern const char __cns11643l1_from_ucs4_tab8[][2];
  92 extern const char __cns11643l1_from_ucs4_tab9[][2];
  93 extern const char __cns11643l1_from_ucs4_tab10[][2];
  94 extern const char __cns11643l1_from_ucs4_tab11[][2];
  95 extern const char __cns11643l1_from_ucs4_tab12[][2];
  96 extern const char __cns11643l1_from_ucs4_tab13[][2];
  97 extern const char __cns11643l1_from_ucs4_tab14[][2];
  98 extern const char __cns11643_from_ucs4_tab[][3];
  99
 100
 101 static inline size_t
 102 ucs4_to_cns11643 (uint32_t wch, char *s, size_t avail)
 103 {
 104   unsigned int ch = (unsigned int) wch;
 105   char buf[2];
 106   const char *cp = buf;
 107   int needed = 2;
 108
 109   switch (ch)
 110     {
 111     case 0xa7 ... 0xf7:
 112       cp = __cns11643l1_from_ucs4_tab1[ch - 0xa7];
 113       break;
 114     case 0x2c7 ... 0x2d9:
 115       cp = __cns11643l1_from_ucs4_tab2[ch - 0x2c7];
 116       break;
 117     case 0x391 ... 0x3c9:
 118       cp = __cns11643l1_from_ucs4_tab3[ch - 0x391];
 119       break;
 120     case 0x2013 ... 0x203e:
 121       cp = __cns11643l1_from_ucs4_tab4[ch - 0x2013];
 122       break;
 123     case 0x2103:
 124       cp = "\x22\x6a";
 125       break;
 126     case 0x2105:
 127       cp = "\x22\x22";
 128       break;
 129     case 0x2109:
 130       cp = "\x22\x6b";
 131       break;
 132     case 0x2160 ... 0x2169:
 133       buf[0] = '\x24';
 134       buf[1] = '\x2b' + (ch - 0x2160);
 135       break;
 136     case 0x2170 ... 0x2179:
 137       buf[0] = '\x26';
 138       buf[1] = '\x35' + (ch - 0x2170);
 139       break;
 140     case 0x2190 ... 0x2199:
 141       cp = __cns11643l1_from_ucs4_tab5[ch - 0x2190];
 142       break;
 143     case 0x2215 ... 0x2267:
 144       cp = __cns11643l1_from_ucs4_tab6[ch - 0x2215];
 145       break;
 146     case 0x22a5:
 147       cp = "\x22\x47";
 148       break;
 149     case 0x22bf:
 150       cp = "\x22\x4a";
 151       break;
 152     case 0x2400 ... 0x2421:
 153       cp = __cns11643l1_from_ucs4_tab7[ch - 0x2400];
 154       break;
 155     case 0x2460 ... 0x247d:
 156       cp = __cns11643l1_from_ucs4_tab8[ch - 0x2460];
 157       break;
 158     case 0x2500 ... 0x2642:
 159       cp = __cns11643l1_from_ucs4_tab9[ch - 0x2500];
 160       break;
 161     case 0x3000 ... 0x3029:
 162       cp = __cns11643l1_from_ucs4_tab10[ch - 0x3000];
 163       break;
 164     case 0x30fb:
 165       cp = "\x21\x26";
 166       break;
 167     case 0x3105 ... 0x3129:
 168       buf[0] = '\x25';
 169       buf[1] = '\x26' + (ch - 0x3105);
 170       break;
 171     case 0x32a3:
 172       cp = "\x22\x21";
 173       break;
 174     case 0x338e ... 0x33d5:
 175       cp = __cns11643l1_from_ucs4_tab11[ch - 0x338e];
 176       break;
 177     case 0x4e00 ... 0x9f9c:
 178       cp = __cns11643l1_from_ucs4_tab12[ch - 0x4e00];
 179
 180       if (cp[0] == '\0')
 181         {
 182           /* Let's try the other planes.  */
 183           needed = 3;
 184           cp = __cns11643_from_ucs4_tab[ch - 0x4e00];
 185         }
 186       break;
 187     case 0xfe30 ... 0xfe6b:
 188       cp = __cns11643l1_from_ucs4_tab13[ch - 0xfe30];
 189       break;
 190     case 0xff01 ... 0xff5d:
 191       cp = __cns11643l1_from_ucs4_tab14[ch - 0xff01];
 192       break;
 193     case 0xffe0:
 194       cp = "\x22\x66";
 195       break;
 196     case 0xffe1:
 197       cp = "\x22\x67";
 198       break;
 199     case 0xffe5:
 200       cp = "\x22\x64";
 201       break;
 202     default:
 203       return __UNKNOWN_10646_CHAR;
 204     }
 205
 206   if (cp[0] == '\0')
 207     return __UNKNOWN_10646_CHAR;
 208
 209   if (avail < needed)
 210     return 0;
 211
 212   s[0] = cp[0];
 213   s[1] = cp[1];
 214   if (needed == 3)
 215     s[2] = cp[2];
 216
 217   return needed;
 218 }