iconvdata/euc-jp.c

   1 /* Mapping tables for EUC-JP 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 <gconv.h>
  23 #include <jis0201.h>
  24 #include <jis0208.h>
  25 #include <jis0212.h>
  26
  27 /* Definitions used in the body of the `gconv' function.  */
  28 #define CHARSET_NAME            "EUC-JP//"
  29 #define FROM_LOOP               from_euc_jp
  30 #define TO_LOOP                 to_euc_jp
  31 #define DEFINE_INIT             1
  32 #define DEFINE_FINI             1
  33 #define MIN_NEEDED_FROM         1
  34 #define MAX_NEEDED_FROM         3
  35 #define MIN_NEEDED_TO           4
  36
  37
  38 /* First define the conversion function from EUC-JP to UCS4.  */
  39 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  40 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  41 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  42 #define LOOPFCT                 FROM_LOOP
  43 #define BODY \
  44   {                                                                           \
  45     uint32_t ch = *inptr;                                                     \
  46                                                                               \
  47     if (ch <= 0x7f)                                                           \
  48       ++inptr;                                                                \
  49     else if ((ch <= 0xa0 || ch > 0xfe) && ch != 0x8e && ch != 0x8f)           \
  50       {                                                                       \
  51         /* This is illegal.  */                                               \
  52         result = __GCONV_ILLEGAL_INPUT;                                       \
  53         break;                                                                \
  54       }                                                                       \
  55     else                                                                      \
  56       {                                                                       \
  57         /* Two or more byte character.  First test whether the next           \
  58            character is also available.  */                                   \
  59         int ch2;                                                              \
  60                                                                               \
  61         if (NEED_LENGTH_TEST && inptr + 1 >= inend)                           \
  62           {                                                                   \
  63             /* The second character is not available.  Store the              \
  64                intermediate result.  */                                       \
  65             result = __GCONV_INCOMPLETE_INPUT;                                \
  66             break;                                                            \
  67           }                                                                   \
  68                                                                               \
  69         ch2 = inptr[1];                                                       \
  70                                                                               \
  71         /* All second bytes of a multibyte character must be >= 0xa1. */      \
  72         if (ch2 < 0xa1)                                                       \
  73           {                                                                   \
  74             /* This is an illegal character.  */                              \
  75             result = __GCONV_ILLEGAL_INPUT;                                   \
  76             break;                                                            \
  77           }                                                                   \
  78                                                                               \
  79         if (ch == 0x8e)                                                       \
  80           {                                                                   \
  81             /* This is code set 2: half-width katakana.  */                   \
  82             ch = jisx0201_to_ucs4 (ch2);                                      \
  83             inptr += 2;                                                       \
  84           }                                                                   \
  85         else                                                                  \
  86           {                                                                   \
  87             const unsigned char *endp;                                        \
  88                                                                               \
  89             if (ch == 0x8f)                                                   \
  90               {                                                               \
  91                 /* This is code set 3: JIS X 0212-1990.  */                   \
  92                 endp = inptr + 1;                                             \
  93                                                                               \
  94                 ch = jisx0212_to_ucs4 (&endp,                                 \
  95                                        NEED_LENGTH_TEST ? inend - endp : 2,   \
  96                                        0x80);                                 \
  97               }                                                               \
  98             else                                                              \
  99               {                                                               \
 100                 /* This is code set 1: JIS X 0208.  */                        \
 101                 endp = inptr;                                                 \
 102                                                                               \
 103                 ch = jisx0208_to_ucs4 (&endp,                                 \
 104                                        NEED_LENGTH_TEST ? inend - inptr : 2,  \
 105                                        0x80);                                 \
 106               }                                                               \
 107                                                                               \
 108             if (NEED_LENGTH_TEST && ch == 0)                                  \
 109               {                                                               \
 110                 /* Not enough input available.  */                            \
 111                 result = __GCONV_INCOMPLETE_INPUT;                            \
 112                 break;                                                        \
 113               }                                                               \
 114             if (ch == __UNKNOWN_10646_CHAR)                                   \
 115               {                                                               \
 116                 /* Illegal character.  */                                     \
 117                 result = __GCONV_ILLEGAL_INPUT;                               \
 118                 break;                                                        \
 119               }                                                               \
 120             inptr = endp;                                                     \
 121           }                                                                   \
 122       }                                                                       \
 123                                                                               \
 124     *((uint32_t *) outptr)++ = ch;                                            \
 125   }
 126 #include <iconv/loop.c>
 127
 128
 129 /* Next, define the other direction.  */
 130 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 131 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 132 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 133 #define LOOPFCT                 TO_LOOP
 134 #define BODY \
 135   {                                                                           \
 136     uint32_t ch = *((uint32_t *) inptr);                                      \
 137                                                                               \
 138     if (ch <= 0x7f)                                                           \
 139       /* It's plain ASCII.  */                                                \
 140       *outptr++ = ch;                                                         \
 141     else if (ch == 0xa5)                                                      \
 142       /* YEN sign => backslash  */                                            \
 143       *outptr++ = 0x5c;                                                       \
 144     else if (ch == 0x203e)                                                    \
 145       /* overscore => asciitilde */                                           \
 146       *outptr++ = 0x7e;                                                       \
 147     else                                                                      \
 148       {                                                                       \
 149         /* Try the JIS character sets.  */                                    \
 150         size_t found;                                                         \
 151                                                                               \
 152         /* See whether we have room for at least two characters.  */          \
 153         if (NEED_LENGTH_TEST && outptr + 1 >= outend)                         \
 154           {                                                                   \
 155             result = __GCONV_FULL_OUTPUT;                                     \
 156             break;                                                            \
 157           }                                                                   \
 158                                                                               \
 159         found = ucs4_to_jisx0201 (ch, outptr + 1);                            \
 160         if (found != __UNKNOWN_10646_CHAR)                                    \
 161           {                                                                   \
 162             /* Yes, it's a JIS 0201 character.  Store the shift byte.  */     \
 163             *outptr = 0x8e;                                                   \
 164             outptr += 2;                                                      \
 165           }                                                                   \
 166         else                                                                  \
 167           {                                                                   \
 168             /* No JIS 0201 character.  */                                     \
 169             found = ucs4_to_jisx0208 (ch, outptr, 2);                         \
 170             /* Please note that we always have enough room for the output. */ \
 171             if (found != __UNKNOWN_10646_CHAR)                                \
 172               {                                                               \
 173                 /* It's a JIS 0208 character, adjust it for EUC-JP.  */       \
 174                 *outptr++ += 0x80;                                            \
 175                 *outptr++ += 0x80;                                            \
 176               }                                                               \
 177             else                                                              \
 178               {                                                               \
 179                 /* No JIS 0208 character.  */                                 \
 180                 found = ucs4_to_jisx0212 (ch, outptr + 1,                     \
 181                                           (NEED_LENGTH_TEST                   \
 182                                            ? outend - outptr - 1 : 2));       \
 183                                                                               \
 184                 if (found == 0)                                               \
 185                   {                                                           \
 186                     /* We ran out of space.  */                               \
 187                     result = __GCONV_FULL_OUTPUT;                             \
 188                     break;                                                    \
 189                   }                                                           \
 190                 else if (found != __UNKNOWN_10646_CHAR)                       \
 191                   {                                                           \
 192                     /* It's a JIS 0212 character, adjust it for EUC-JP.  */   \
 193                     *outptr++ = 0x8f;                                         \
 194                     *outptr++ += 0x80;                                        \
 195                     *outptr++ += 0x80;                                        \
 196                   }                                                           \
 197                 else                                                          \
 198                   {                                                           \
 199                     /* Illegal character.  */                                 \
 200                     result = __GCONV_ILLEGAL_INPUT;                           \
 201                     break;                                                    \
 202                   }                                                           \
 203               }                                                               \
 204           }                                                                   \
 205       }                                                                       \
 206                                                                               \
 207     inptr += 4;                                                               \
 208   }
 209 #include <iconv/loop.c>
 210
 211
 212 /* Now define the toplevel functions.  */
 213 #include <iconv/skeleton.c>