iconvdata/euc-tw.c

   1 /* Mapping tables for EUC-TW 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 #include <cns11643l1.h>
  23 #include <cns11643.h>
  24
  25 /* Definitions used in the body of the `gconv' function.  */
  26 #define CHARSET_NAME            "EUC-TW//"
  27 #define FROM_LOOP               from_euc_tw
  28 #define TO_LOOP                 to_euc_tw
  29 #define DEFINE_INIT             1
  30 #define DEFINE_FINI             1
  31 #define MIN_NEEDED_FROM         1
  32 #define MAX_NEEDED_FROM         4
  33 #define MIN_NEEDED_TO           4
  34
  35
  36 /* First define the conversion function from EUC-TW to UCS4.  */
  37 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  38 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  39 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  40 #define LOOPFCT                 FROM_LOOP
  41 #define BODY \
  42   {                                                                           \
  43     uint32_t ch = *inptr;                                                     \
  44                                                                               \
  45     if (ch <= 0x7f)                                                           \
  46       /* Plain ASCII.  */                                                     \
  47       ++inptr;                                                                \
  48     else if ((ch <= 0xa0 || ch > 0xfe) && ch != 0x8e)                         \
  49       {                                                                       \
  50         /* This is illegal.  */                                               \
  51         result = __GCONV_ILLEGAL_INPUT;                                       \
  52         break;                                                                \
  53       }                                                                       \
  54     else                                                                      \
  55       {                                                                       \
  56         /* Two or more byte character.  First test whether the next           \
  57            character is also available.  */                                   \
  58         uint32_t ch2;                                                         \
  59                                                                               \
  60         if (NEED_LENGTH_TEST && inptr + (ch == 0x8e ? 3 : 1) >= inend)        \
  61           {                                                                   \
  62             /* The second character is not available.  Store the              \
  63                intermediate result.  */                                       \
  64             result = __GCONV_INCOMPLETE_INPUT;                                \
  65             break;                                                            \
  66           }                                                                   \
  67                                                                               \
  68         ch2 = *inptr;                                                         \
  69                                                                               \
  70         /* All second bytes of a multibyte character must be >= 0xa1. */      \
  71         if (ch2 < 0xa1 || ch2 == 0xff)                                        \
  72           {                                                                   \
  73             /* This is an illegal character.  */                              \
  74             result = __GCONV_ILLEGAL_INPUT;                                   \
  75             break;                                                            \
  76           }                                                                   \
  77                                                                               \
  78         if (ch == 0x8e)                                                       \
  79           {                                                                   \
  80             /* This is code set 2: CNS 11643, planes 1 to 16.  */             \
  81             const char *endp = inptr + 1;                                     \
  82                                                                               \
  83             ch = cns11643_to_ucs4 (&endp,                                     \
  84                                    NEED_LENGTH_TEST ? inend - inptr - 1 : 3,  \
  85                                    0x80);                                     \
  86             /* Please note that we need not test for the missing input        \
  87                characters here anymore.  */                                   \
  88             if (ch == __UNKNOWN_10646_CHAR)                                   \
  89               {                                                               \
  90                 /* Illegal input.  */                                         \
  91                 result = __GCONV_ILLEGAL_INPUT;                               \
  92                 break;                                                        \
  93               }                                                               \
  94                                                                               \
  95             inptr += 4;                                                       \
  96           }                                                                   \
  97         else                                                                  \
  98           {                                                                   \
  99             /* This is code set 1: CNS 11643, plane 1.  */                    \
 100             const char *endp = inptr;                                         \
 101                                                                               \
 102             ch = cns11643l1_to_ucs4 (&endp,                                   \
 103                                      NEED_LENGTH_TEST ? inend - inptr : 2,    \
 104                                      0x80);                                   \
 105             /* Please note that we need not test for the missing input        \
 106                characters here anymore.  */                                   \
 107             if (ch == __UNKNOWN_10646_CHAR)                                   \
 108               {                                                               \
 109                 /* Illegal input.  */                                         \
 110                 result = __GCONV_ILLEGAL_INPUT;                               \
 111                 break;                                                        \
 112               }                                                               \
 113                                                                               \
 114             inptr += 2;                                                       \
 115           }                                                                   \
 116       }                                                                       \
 117                                                                               \
 118     *((uint32_t *) outptr)++ = ch;                                            \
 119   }
 120 #include <iconv/loop.c>
 121
 122
 123 /* Next, define the other direction.  */
 124 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 125 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 126 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 127 #define LOOPFCT                 TO_LOOP
 128 #define BODY \
 129   {                                                                           \
 130     uint32_t ch = *((uint32_t *) inptr);                                      \
 131                                                                               \
 132     if (ch <= 0x7f)                                                           \
 133       /* It's plain ASCII.  */                                                \
 134       *outptr++ = ch;                                                         \
 135     else                                                                      \
 136       {                                                                       \
 137         /* Try the JIS character sets.  */                                    \
 138         size_t found;                                                         \
 139                                                                               \
 140         found = ucs4_to_cns11643l1 (ch, outptr,                               \
 141                                     NEED_LENGTH_TEST ? outend - outptr : 2);  \
 142         if (NEED_LENGTH_TEST && found == 0)                                   \
 143           {                                                                   \
 144             /* We ran out of space.  */                                       \
 145             result = __GCONV_INCOMPLETE_INPUT;                                \
 146             break;                                                            \
 147           }                                                                   \
 148         if (found != __UNKNOWN_10646_CHAR)                                    \
 149           {                                                                   \
 150             /* It's a CNS 11643, plane 1 character, adjust it for EUC-TW.  */ \
 151             *outptr++ += 0x80;                                                \
 152             *outptr++ += 0x80;                                                \
 153           }                                                                   \
 154         else                                                                  \
 155           {                                                                   \
 156             /* No CNS 11643, plane 1 character.  */                           \
 157                                                                               \
 158             found = ucs4_to_cns11643 (ch, outptr + 1,                         \
 159                                       (NEED_LENGTH_TEST                       \
 160                                        ? outend - outptr - 1 : 3));           \
 161             if (NEED_LENGTH_TEST && found == 0)                               \
 162               {                                                               \
 163                 /* We ran out of space.  */                                   \
 164                 result = __GCONV_INCOMPLETE_INPUT;                            \
 165                 break;                                                        \
 166               }                                                               \
 167             if (found == __UNKNOWN_10646_CHAR)                                \
 168               {                                                               \
 169                 /* No legal input.  */                                        \
 170                 result = __GCONV_ILLEGAL_INPUT;                               \
 171                 break;                                                        \
 172               }                                                               \
 173                                                                               \
 174             /* It's a CNS 11643 character, adjust it for EUC-TW.  */          \
 175             *outptr++ = '\x8e';                                               \
 176             *outptr++ += 0xa0;                                                \
 177             *outptr++ += 0x80;                                                \
 178             *outptr++ += 0x80;                                                \
 179           }                                                                   \
 180       }                                                                       \
 181                                                                               \
 182     inptr += 4;                                                               \
 183   }
 184 #include <iconv/loop.c>
 185
 186
 187 /* Now define the toplevel functions.  */
 188 #include <iconv/skeleton.c>