iconvdata/cns11643.h

   1 /* Access functions for CNS 11643, plane 2 handling.
   2    Copyright (C) 1998, 1999, 2000 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 1 to UCS4 conversion.  */
  24 extern const uint16_t __cns11643l1_to_ucs4_tab[];
  25 /* Table for CNS 11643, plane 2 to UCS4 conversion.  */
  26 extern const uint16_t __cns11643l2_to_ucs4_tab[];
  27 /* Table for CNS 11643, plane 14 to UCS4 conversion.  */
  28 extern const uint16_t __cns11643l14_to_ucs4_tab[];
  29
  30
  31 static inline uint32_t
  32 cns11643_to_ucs4 (const char **s, size_t avail, unsigned char offset)
  33 {
  34   unsigned char ch = *(*s);
  35   unsigned char ch2;
  36   unsigned char ch3;
  37   uint32_t result;
  38   int idx;
  39
  40   if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) > 0x30)
  41     return __UNKNOWN_10646_CHAR;
  42
  43   if (avail < 3)
  44     return 0;
  45
  46   ch2 = (*s)[1];
  47   if ((ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
  48     return __UNKNOWN_10646_CHAR;
  49
  50   ch3 = (*s)[2];
  51   if ((ch3 - offset) <= 0x20 || (ch3 - offset) >= 0x7f)
  52     return __UNKNOWN_10646_CHAR;
  53
  54   idx = (ch2 - 0x21 - offset) * 94 + (ch3 - 0x21 - offset);
  55
  56   if ((ch - 0x20 - offset) == 1)
  57     {
  58       if (idx > 0x21f2)
  59         return __UNKNOWN_10646_CHAR;
  60       result = __cns11643l1_to_ucs4_tab[idx];
  61     }
  62   else if ((ch - 0x20 - offset) == 2)
  63     {
  64       if (idx > 0x1de1)
  65         return __UNKNOWN_10646_CHAR;
  66       result = __cns11643l2_to_ucs4_tab[idx];
  67     }
  68   else if ((ch - 0x20 - offset) == 0xe)
  69     {
  70       if (idx > 0x19bd)
  71         return __UNKNOWN_10646_CHAR;
  72       result = __cns11643l14_to_ucs4_tab[idx];
  73     }
  74   else
  75     return __UNKNOWN_10646_CHAR;
  76
  77   if (result != L'\0')
  78     (*s) += 3;
  79   else
  80     result = __UNKNOWN_10646_CHAR;
  81
  82   return result;
  83 }
  84
  85
  86 /* Tables for the UCS4 -> CNS conversion.  */
  87 extern const char __cns11643l1_from_ucs4_tab1[][2];
  88 extern const char __cns11643l1_from_ucs4_tab2[][2];
  89 extern const char __cns11643l1_from_ucs4_tab3[][2];
  90 extern const char __cns11643l1_from_ucs4_tab4[][2];
  91 extern const char __cns11643l1_from_ucs4_tab5[][2];
  92 extern const char __cns11643l1_from_ucs4_tab6[][2];
  93 extern const char __cns11643l1_from_ucs4_tab7[][2];
  94 extern const char __cns11643l1_from_ucs4_tab8[][2];
  95 extern const char __cns11643l1_from_ucs4_tab9[][2];
  96 extern const char __cns11643l1_from_ucs4_tab10[][2];
  97 extern const char __cns11643l1_from_ucs4_tab11[][2];
  98 extern const char __cns11643l1_from_ucs4_tab12[][2];
  99 extern const char __cns11643l1_from_ucs4_tab13[][2];
 100 extern const char __cns11643l1_from_ucs4_tab14[][2];
 101 extern const char __cns11643_from_ucs4_tab[][3];
 102
 103
 104 static inline size_t
 105 ucs4_to_cns11643 (uint32_t wch, char *s, size_t avail)
 106 {
 107   unsigned int ch = (unsigned int) wch;
 108   char buf[2];
 109   const char *cp = buf;
 110   int needed = 2;
 111
 112   switch (ch)
 113     {
 114     case 0xa7 ... 0xf7:
 115       cp = __cns11643l1_from_ucs4_tab1[ch - 0xa7];
 116       break;
 117     case 0x2c7 ... 0x2d9:
 118       cp = __cns11643l1_from_ucs4_tab2[ch - 0x2c7];
 119       break;
 120     case 0x391 ... 0x3c9:
 121       cp = __cns11643l1_from_ucs4_tab3[ch - 0x391];
 122       break;
 123     case 0x2013 ... 0x203e:
 124       cp = __cns11643l1_from_ucs4_tab4[ch - 0x2013];
 125       break;
 126     case 0x2103:
 127       cp = "\x22\x6a";
 128       break;
 129     case 0x2105:
 130       cp = "\x22\x22";
 131       break;
 132     case 0x2109:
 133       cp = "\x22\x6b";
 134       break;
 135     case 0x2160 ... 0x2169:
 136       buf[0] = '\x24';
 137       buf[1] = '\x2b' + (ch - 0x2160);
 138       break;
 139     case 0x2170 ... 0x2179:
 140       buf[0] = '\x26';
 141       buf[1] = '\x35' + (ch - 0x2170);
 142       break;
 143     case 0x2190 ... 0x2199:
 144       cp = __cns11643l1_from_ucs4_tab5[ch - 0x2190];
 145       break;
 146     case 0x2215 ... 0x2267:
 147       cp = __cns11643l1_from_ucs4_tab6[ch - 0x2215];
 148       break;
 149     case 0x22a5:
 150       cp = "\x22\x47";
 151       break;
 152     case 0x22bf:
 153       cp = "\x22\x4a";
 154       break;
 155     case 0x2400 ... 0x2421:
 156       cp = __cns11643l1_from_ucs4_tab7[ch - 0x2400];
 157       break;
 158     case 0x2460 ... 0x247d:
 159       cp = __cns11643l1_from_ucs4_tab8[ch - 0x2460];
 160       break;
 161     case 0x2500 ... 0x2642:
 162       cp = __cns11643l1_from_ucs4_tab9[ch - 0x2500];
 163       break;
 164     case 0x3000 ... 0x3029:
 165       cp = __cns11643l1_from_ucs4_tab10[ch - 0x3000];
 166       break;
 167     case 0x30fb:
 168       cp = "\x21\x26";
 169       break;
 170     case 0x3105 ... 0x3129:
 171       buf[0] = '\x25';
 172       buf[1] = '\x47' + (ch - 0x3105);
 173       break;
 174     case 0x32a3:
 175       cp = "\x22\x21";
 176       break;
 177     case 0x338e ... 0x33d5:
 178       cp = __cns11643l1_from_ucs4_tab11[ch - 0x338e];
 179       break;
 180     case 0x4e00 ... 0x9f9c:
 181       cp = __cns11643l1_from_ucs4_tab12[ch - 0x4e00];
 182
 183       if (cp[0] != '\0')
 184         break;
 185       /* FALLTHROUGH.  Let's try the other planes.  */
 186     case 0x9f9d ... 0x9fa5:
 187       /* Let's try the other planes.  */
 188       needed = 3;
 189       cp = __cns11643_from_ucs4_tab[ch - 0x4e00];
 190       break;
 191     case 0xfe30 ... 0xfe6b:
 192       cp = __cns11643l1_from_ucs4_tab13[ch - 0xfe30];
 193       break;
 194     case 0xff01 ... 0xff5d:
 195       cp = __cns11643l1_from_ucs4_tab14[ch - 0xff01];
 196       break;
 197     case 0xffe0:
 198       cp = "\x22\x66";
 199       break;
 200     case 0xffe1:
 201       cp = "\x22\x67";
 202       break;
 203     case 0xffe5:
 204       cp = "\x22\x64";
 205       break;
 206     default:
 207       return __UNKNOWN_10646_CHAR;
 208     }
 209
 210   if (cp[0] == '\0')
 211     return __UNKNOWN_10646_CHAR;
 212
 213   if (avail < needed)
 214     return 0;
 215
 216   s[0] = cp[0];
 217   s[1] = cp[1];
 218   if (needed == 3)
 219     s[2] = cp[2];
 220
 221   return needed;
 222 }