iconvdata/euc-jp.c

   1 /* Mapping tables for EUC-JP 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 <dlfcn.h>
  22 #include <stdint.h>
  23 #include <gconv.h>
  24 #include <jis0201.h>
  25 #include <jis0208.h>
  26 #include <jis0212.h>
  27
  28 /* Definitions used in the body of the `gconv' function.  */
  29 #define CHARSET_NAME            "EUC-JP//"
  30 #define FROM_LOOP               from_euc_jp
  31 #define TO_LOOP                 to_euc_jp
  32 #define DEFINE_INIT             1
  33 #define DEFINE_FINI             1
  34 #define MIN_NEEDED_FROM         1
  35 #define MAX_NEEDED_FROM         3
  36 #define MIN_NEEDED_TO           4
  37
  38
  39 /* First define the conversion function from EUC-JP to UCS4.  */
  40 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  41 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  42 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  43 #define LOOPFCT                 FROM_LOOP
  44 #define BODY \
  45   {                                                                           \
  46     uint32_t ch = *inptr;                                                     \
  47                                                                               \
  48     if (ch <= 0x7f)                                                           \
  49       ++inptr;                                                                \
  50     else if ((__builtin_expect (ch, 0xa1) <= 0xa0 && ch != 0x8e && ch != 0x8f)\
  51              || __builtin_expect (ch, 0xfe) > 0xfe)                           \
  52       {                                                                       \
  53         /* This is illegal.  */                                               \
  54         if (! ignore_errors_p ())                                             \
  55           {                                                                   \
  56             result = __GCONV_ILLEGAL_INPUT;                                   \
  57             break;                                                            \
  58           }                                                                   \
  59                                                                               \
  60         ++inptr;                                                              \
  61         ++*irreversible;                                                      \
  62         continue;                                                             \
  63       }                                                                       \
  64     else                                                                      \
  65       {                                                                       \
  66         /* Two or more byte character.  First test whether the next           \
  67            character is also available.  */                                   \
  68         int ch2;                                                              \
  69                                                                               \
  70         if (__builtin_expect (inptr + 1 >= inend, 0))                         \
  71           {                                                                   \
  72             /* The second character is not available.  Store the              \
  73                intermediate result.  */                                       \
  74             result = __GCONV_INCOMPLETE_INPUT;                                \
  75             break;                                                            \
  76           }                                                                   \
  77                                                                               \
  78         ch2 = inptr[1];                                                       \
  79                                                                               \
  80         /* All second bytes of a multibyte character must be >= 0xa1. */      \
  81         if (__builtin_expect (ch2, 0xa1) < 0xa1)                              \
  82           {                                                                   \
  83             /* This is an illegal character.  */                              \
  84             if (! ignore_errors_p ())                                         \
  85               {                                                               \
  86                 result = __GCONV_ILLEGAL_INPUT;                               \
  87                 break;                                                        \
  88               }                                                               \
  89                                                                               \
  90             ++inptr;                                                          \
  91             ++*irreversible;                                                  \
  92             continue;                                                         \
  93           }                                                                   \
  94                                                                               \
  95         if (ch == 0x8e)                                                       \
  96           {                                                                   \
  97             /* This is code set 2: half-width katakana.  */                   \
  98             ch = jisx0201_to_ucs4 (ch2);                                      \
  99             inptr += 2;                                                       \
 100           }                                                                   \
 101         else                                                                  \
 102           {                                                                   \
 103             const unsigned char *endp;                                        \
 104                                                                               \
 105             if (ch == 0x8f)                                                   \
 106               {                                                               \
 107                 /* This is code set 3: JIS X 0212-1990.  */                   \
 108                 endp = inptr + 1;                                             \
 109                                                                               \
 110                 ch = jisx0212_to_ucs4 (&endp, inend - endp, 0x80);            \
 111               }                                                               \
 112             else                                                              \
 113               {                                                               \
 114                 /* This is code set 1: JIS X 0208.  */                        \
 115                 endp = inptr;                                                 \
 116                                                                               \
 117                 ch = jisx0208_to_ucs4 (&endp, inend - inptr, 0x80);           \
 118               }                                                               \
 119                                                                               \
 120             if (__builtin_expect (ch, 1) == 0)                                \
 121               {                                                               \
 122                 /* Not enough input available.  */                            \
 123                 result = __GCONV_INCOMPLETE_INPUT;                            \
 124                 break;                                                        \
 125               }                                                               \
 126             if (__builtin_expect (ch, 0) == __UNKNOWN_10646_CHAR)             \
 127               {                                                               \
 128                 /* Illegal character.  */                                     \
 129                 if (! ignore_errors_p ())                                     \
 130                   {                                                           \
 131                     /* This is an illegal character.  */                      \
 132                     result = __GCONV_ILLEGAL_INPUT;                           \
 133                     break;                                                    \
 134                   }                                                           \
 135                                                                               \
 136                 inptr += 2;                                                   \
 137                 ++*irreversible;                                              \
 138                 continue;                                                     \
 139               }                                                               \
 140             inptr = endp;                                                     \
 141           }                                                                   \
 142       }                                                                       \
 143                                                                               \
 144     put32 (outptr, ch);                                                       \
 145     outptr += 4;                                                              \
 146   }
 147 #define LOOP_NEED_FLAGS
 148 #include <iconv/loop.c>
 149
 150
 151 /* Next, define the other direction.  */
 152 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 153 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 154 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 155 #define LOOPFCT                 TO_LOOP
 156 #define BODY \
 157   {                                                                           \
 158     uint32_t ch = get32 (inptr);                                              \
 159                                                                               \
 160     if (ch <= 0x7f)                                                           \
 161       /* It's plain ASCII.  */                                                \
 162       *outptr++ = ch;                                                         \
 163     else if (ch == 0xa5)                                                      \
 164       /* YEN sign => backslash  */                                            \
 165       *outptr++ = 0x5c;                                                       \
 166     else if (ch == 0x203e)                                                    \
 167       /* overscore => asciitilde */                                           \
 168       *outptr++ = 0x7e;                                                       \
 169     else                                                                      \
 170       {                                                                       \
 171         /* Try the JIS character sets.  */                                    \
 172         size_t found;                                                         \
 173                                                                               \
 174         /* See whether we have room for at least two characters.  */          \
 175         if (__builtin_expect (outptr + 1 >= outend, 0))                       \
 176           {                                                                   \
 177             result = __GCONV_FULL_OUTPUT;                                     \
 178             break;                                                            \
 179           }                                                                   \
 180                                                                               \
 181         found = ucs4_to_jisx0201 (ch, outptr + 1);                            \
 182         if (found != __UNKNOWN_10646_CHAR)                                    \
 183           {                                                                   \
 184             /* Yes, it's a JIS 0201 character.  Store the shift byte.  */     \
 185             *outptr = 0x8e;                                                   \
 186             outptr += 2;                                                      \
 187           }                                                                   \
 188         else                                                                  \
 189           {                                                                   \
 190             /* No JIS 0201 character.  */                                     \
 191             found = ucs4_to_jisx0208 (ch, outptr, 2);                         \
 192             /* Please note that we always have enough room for the output. */ \
 193             if (found != __UNKNOWN_10646_CHAR)                                \
 194               {                                                               \
 195                 /* It's a JIS 0208 character, adjust it for EUC-JP.  */       \
 196                 *outptr++ += 0x80;                                            \
 197                 *outptr++ += 0x80;                                            \
 198               }                                                               \
 199             else                                                              \
 200               {                                                               \
 201                 /* No JIS 0208 character.  */                                 \
 202                 found = ucs4_to_jisx0212 (ch, outptr + 1,                     \
 203                                           outend - outptr - 1);               \
 204                                                                               \
 205                 if (__builtin_expect (found, 1) == 0)                         \
 206                   {                                                           \
 207                     /* We ran out of space.  */                               \
 208                     result = __GCONV_FULL_OUTPUT;                             \
 209                     break;                                                    \
 210                   }                                                           \
 211                 else if (__builtin_expect (found, 0) != __UNKNOWN_10646_CHAR) \
 212                   {                                                           \
 213                     /* It's a JIS 0212 character, adjust it for EUC-JP.  */   \
 214                     *outptr++ = 0x8f;                                         \
 215                     *outptr++ += 0x80;                                        \
 216                     *outptr++ += 0x80;                                        \
 217                   }                                                           \
 218                 else                                                          \
 219                   {                                                           \
 220                     /* Illegal character.  */                                 \
 221                     if (step_data->__trans.__trans_fct != NULL)               \
 222                       {                                                       \
 223                         result = DL_CALL_FCT (step_data->__trans.__trans_fct, \
 224                                               (step, step_data, *inptrp,      \
 225                                                &inptr, inend, *outptrp,       \
 226                                                &outptr, outend,               \
 227                                                irreversible));                \
 228                         if (result != __GCONV_OK)                             \
 229                           break;                                              \
 230                       }                                                       \
 231                     else if (! ignore_errors_p ())                            \
 232                       {                                                       \
 233                         result = __GCONV_ILLEGAL_INPUT;                       \
 234                         break;                                                \
 235                       }                                                       \
 236                     else                                                      \
 237                       {                                                       \
 238                         inptr += 4;                                           \
 239                         ++*irreversible;                                      \
 240                       }                                                       \
 241                     continue;                                                 \
 242                   }                                                           \
 243               }                                                               \
 244           }                                                                   \
 245       }                                                                       \
 246                                                                               \
 247     inptr += 4;                                                               \
 248   }
 249 #define LOOP_NEED_FLAGS
 250 #include <iconv/loop.c>
 251
 252
 253 /* Now define the toplevel functions.  */
 254 #include <iconv/skeleton.c>