iconvdata/euc-tw.c

   1 /* Mapping tables for EUC-TW handling.
   2    Copyright (C) 1998-2023 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, see
  17    <https://www.gnu.org/licenses/>.  */
  18
  19 #include <dlfcn.h>
  20 #include <stdint.h>
  21 #include <cns11643l1.h>
  22 #include <cns11643.h>
  23
  24 /* Definitions used in the body of the `gconv' function.  */
  25 #define CHARSET_NAME            "EUC-TW//"
  26 #define FROM_LOOP               from_euc_tw
  27 #define TO_LOOP                 to_euc_tw
  28 #define DEFINE_INIT             1
  29 #define DEFINE_FINI             1
  30 #define MIN_NEEDED_FROM         1
  31 #define MAX_NEEDED_FROM         4
  32 #define MIN_NEEDED_TO           4
  33 #define ONE_DIRECTION           0
  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         STANDARD_FROM_LOOP_ERR_HANDLER (1);                                   \
  52       }                                                                       \
  53     else                                                                      \
  54       {                                                                       \
  55         /* Two or more byte character.  First test whether the next byte      \
  56            is also available.  */                                             \
  57         uint32_t ch2;                                                         \
  58                                                                               \
  59         if (inptr + 1 >= inend)                                               \
  60           {                                                                   \
  61             /* The second byte is not available.  Store the intermediate      \
  62                result.  */                                                    \
  63             result = __GCONV_INCOMPLETE_INPUT;                                \
  64             break;                                                            \
  65           }                                                                   \
  66                                                                               \
  67         ch2 = *(inptr + 1);                                                   \
  68                                                                               \
  69         /* All second bytes of a multibyte character must be >= 0xa1. */      \
  70         if (ch2 < 0xa1 || ch2 == 0xff)                                        \
  71           STANDARD_FROM_LOOP_ERR_HANDLER (1);                                 \
  72                                                                               \
  73         if (ch == 0x8e)                                                       \
  74           {                                                                   \
  75             /* This is code set 2: CNS 11643, planes 1 to 16.  */             \
  76             const unsigned char *endp = inptr + 1;                            \
  77                                                                               \
  78             ch = cns11643_to_ucs4 (&endp, inend - inptr - 1, 0x80);           \
  79                                                                               \
  80             if (ch == 0)                                                      \
  81               {                                                               \
  82                 /* The third or fourth byte is not available.  Store          \
  83                    the intermediate result.  */                               \
  84                 result = __GCONV_INCOMPLETE_INPUT;                            \
  85                 break;                                                        \
  86               }                                                               \
  87                                                                               \
  88             if (ch == __UNKNOWN_10646_CHAR)                                   \
  89               /* Illegal input.  */                                           \
  90               STANDARD_FROM_LOOP_ERR_HANDLER (1);                             \
  91                                                                               \
  92             inptr += 4;                                                       \
  93           }                                                                   \
  94         else                                                                  \
  95           {                                                                   \
  96             /* This is code set 1: CNS 11643, plane 1.  */                    \
  97             const unsigned char *endp = inptr;                                \
  98                                                                               \
  99             ch = cns11643l1_to_ucs4 (&endp, inend - inptr, 0x80);             \
 100             /* Please note that we need not test for the missing input        \
 101                characters here anymore.  */                                   \
 102             if (ch == __UNKNOWN_10646_CHAR)                                   \
 103               /* Illegal input.  */                                           \
 104               STANDARD_FROM_LOOP_ERR_HANDLER (2);                             \
 105                                                                               \
 106             inptr += 2;                                                       \
 107           }                                                                   \
 108       }                                                                       \
 109                                                                               \
 110     put32 (outptr, ch);                                                       \
 111     outptr += 4;                                                              \
 112   }
 113 #define LOOP_NEED_FLAGS
 114 #define ONEBYTE_BODY \
 115   {                                                                           \
 116     if (c < 0x80)                                                             \
 117       return c;                                                               \
 118     else                                                                      \
 119       return WEOF;                                                            \
 120   }
 121 #include <iconv/loop.c>
 122
 123
 124 /* Next, define the other direction.  */
 125 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 126 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 127 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 128 #define LOOPFCT                 TO_LOOP
 129 #define BODY \
 130   {                                                                           \
 131     uint32_t ch = get32 (inptr);                                              \
 132                                                                               \
 133     if (ch <= 0x7f)                                                           \
 134       /* It's plain ASCII.  */                                                \
 135       *outptr++ = ch;                                                         \
 136     else                                                                      \
 137       {                                                                       \
 138         /* Try the CNS 11643 planes.  */                                      \
 139         size_t found;                                                         \
 140                                                                               \
 141         found = ucs4_to_cns11643l1 (ch, outptr, outend - outptr);             \
 142         if (__builtin_expect (found, 1) == 0)                                 \
 143           {                                                                   \
 144             /* We ran out of space.  */                                       \
 145             result = __GCONV_FULL_OUTPUT;                                     \
 146             break;                                                            \
 147           }                                                                   \
 148         if (__builtin_expect (found, 1) != __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, outend - outptr - 1);   \
 159             if (__builtin_expect (found, 1) == 0)                             \
 160               {                                                               \
 161                 /* We ran out of space.  */                                   \
 162                 result = __GCONV_FULL_OUTPUT;                                 \
 163                 break;                                                        \
 164               }                                                               \
 165             if (__builtin_expect (found, 0) == __UNKNOWN_10646_CHAR)          \
 166               {                                                               \
 167                 UNICODE_TAG_HANDLER (ch, 4);                                  \
 168                                                                               \
 169                 /* Illegal character.  */                                     \
 170                 STANDARD_TO_LOOP_ERR_HANDLER (4);                             \
 171               }                                                               \
 172                                                                               \
 173             /* It's a CNS 11643 character, adjust it for EUC-TW.  */          \
 174             *outptr++ = '\x8e';                                               \
 175             *outptr++ += 0xa0;                                                \
 176             *outptr++ += 0x80;                                                \
 177             *outptr++ += 0x80;                                                \
 178           }                                                                   \
 179       }                                                                       \
 180                                                                               \
 181     inptr += 4;                                                               \
 182   }
 183 #define LOOP_NEED_FLAGS
 184 #include <iconv/loop.c>
 185
 186
 187 /* Now define the toplevel functions.  */
 188 #include <iconv/skeleton.c>