iconvdata/euc-jp.c

   1 /* Mapping tables for EUC-JP 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 <gconv.h>
  22 #include <jis0201.h>
  23 #include <jis0208.h>
  24 #include <jis0212.h>
  25
  26 /* Definitions used in the body of the `gconv' function.  */
  27 #define CHARSET_NAME            "EUC-JP//"
  28 #define FROM_LOOP               from_euc_jp
  29 #define TO_LOOP                 to_euc_jp
  30 #define DEFINE_INIT             1
  31 #define DEFINE_FINI             1
  32 #define MIN_NEEDED_FROM         1
  33 #define MAX_NEEDED_FROM         3
  34 #define MIN_NEEDED_TO           4
  35 #define ONE_DIRECTION           0
  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 < 0x8e || (ch >= 0x90 && ch <= 0x9f))                              \
  48       ++inptr;                                                                \
  49     else if (ch == 0xff)                                                      \
  50       {                                                                       \
  51         /* This is illegal.  */                                               \
  52         STANDARD_FROM_LOOP_ERR_HANDLER (1);                                   \
  53       }                                                                       \
  54     else                                                                      \
  55       {                                                                       \
  56         /* Two or more byte character.  First test whether the next           \
  57            byte is also available.  */                                        \
  58         int ch2;                                                              \
  59                                                                               \
  60         if (__glibc_unlikely (inptr + 1 >= inend))                            \
  61           {                                                                   \
  62             /* The second byte is not available.  Store the                   \
  63                intermediate result.  */                                       \
  64             result = __GCONV_INCOMPLETE_INPUT;                                \
  65             break;                                                            \
  66           }                                                                   \
  67                                                                               \
  68         ch2 = inptr[1];                                                       \
  69                                                                               \
  70         /* All second bytes of a multibyte character must be >= 0xa1. */      \
  71         if (__glibc_unlikely (ch2 < 0xa1))                                    \
  72           STANDARD_FROM_LOOP_ERR_HANDLER (1);                                 \
  73                                                                               \
  74         if (ch == 0x8e)                                                       \
  75           {                                                                   \
  76             /* This is code set 2: half-width katakana.  */                   \
  77             ch = jisx0201_to_ucs4 (ch2);                                      \
  78             if (__builtin_expect (ch, 0) == __UNKNOWN_10646_CHAR)             \
  79               STANDARD_FROM_LOOP_ERR_HANDLER (1);                             \
  80                                                                               \
  81             inptr += 2;                                                       \
  82           }                                                                   \
  83         else                                                                  \
  84           {                                                                   \
  85             const unsigned char *endp;                                        \
  86                                                                               \
  87             if (ch == 0x8f)                                                   \
  88               {                                                               \
  89                 /* This is code set 3: JIS X 0212-1990.  */                   \
  90                 endp = inptr + 1;                                             \
  91                                                                               \
  92                 ch = jisx0212_to_ucs4 (&endp, inend - endp, 0x80);            \
  93               }                                                               \
  94             else                                                              \
  95               {                                                               \
  96                 /* This is code set 1: JIS X 0208.  */                        \
  97                 endp = inptr;                                                 \
  98                                                                               \
  99                 ch = jisx0208_to_ucs4 (&endp, inend - inptr, 0x80);           \
 100               }                                                               \
 101                                                                               \
 102             if (__builtin_expect (ch, 1) == 0)                                \
 103               {                                                               \
 104                 /* Not enough input available.  */                            \
 105                 result = __GCONV_INCOMPLETE_INPUT;                            \
 106                 break;                                                        \
 107               }                                                               \
 108             if (__glibc_unlikely (ch == __UNKNOWN_10646_CHAR))                \
 109               /* Illegal character.  */                                       \
 110               STANDARD_FROM_LOOP_ERR_HANDLER (1);                             \
 111                                                                               \
 112             inptr = endp;                                                     \
 113           }                                                                   \
 114       }                                                                       \
 115                                                                               \
 116     put32 (outptr, ch);                                                       \
 117     outptr += 4;                                                              \
 118   }
 119 #define ONEBYTE_BODY \
 120   {                                                                           \
 121     if (c < 0x8e || (c >= 0x90 && c <= 0x9f))                                 \
 122       return c;                                                               \
 123     else                                                                      \
 124       return WEOF;                                                            \
 125   }
 126 #define LOOP_NEED_FLAGS
 127 #include <iconv/loop.c>
 128
 129
 130 /* Next, define the other direction.  */
 131 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 132 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 133 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 134 #define LOOPFCT                 TO_LOOP
 135 #define BODY \
 136   {                                                                           \
 137     uint32_t ch = get32 (inptr);                                              \
 138                                                                               \
 139     if (ch < 0x8e || (ch >= 0x90 && ch <= 0x9f))                              \
 140       /* It's plain ASCII or C1.  */                                          \
 141       *outptr++ = ch;                                                         \
 142     else if (ch == 0xa5)                                                      \
 143       /* YEN sign => backslash  */                                            \
 144       *outptr++ = 0x5c;                                                       \
 145     else if (ch == 0x203e)                                                    \
 146       /* overscore => asciitilde */                                           \
 147       *outptr++ = 0x7e;                                                       \
 148     else                                                                      \
 149       {                                                                       \
 150         /* Try the JIS character sets.  */                                    \
 151         size_t found;                                                         \
 152                                                                               \
 153         /* See whether we have room for at least two characters.  */          \
 154         if (__glibc_unlikely (outptr + 1 >= outend))                          \
 155           {                                                                   \
 156             result = __GCONV_FULL_OUTPUT;                                     \
 157             break;                                                            \
 158           }                                                                   \
 159                                                                               \
 160         found = ucs4_to_jisx0201 (ch, outptr + 1);                            \
 161         if (found != __UNKNOWN_10646_CHAR)                                    \
 162           {                                                                   \
 163             /* Yes, it's a JIS 0201 character.  Store the shift byte.  */     \
 164             *outptr = 0x8e;                                                   \
 165             outptr += 2;                                                      \
 166           }                                                                   \
 167         else                                                                  \
 168           {                                                                   \
 169             /* No JIS 0201 character.  */                                     \
 170             found = ucs4_to_jisx0208 (ch, outptr, 2);                         \
 171             /* Please note that we always have enough room for the output. */ \
 172             if (found != __UNKNOWN_10646_CHAR)                                \
 173               {                                                               \
 174                 /* It's a JIS 0208 character, adjust it for EUC-JP.  */       \
 175                 *outptr++ += 0x80;                                            \
 176                 *outptr++ += 0x80;                                            \
 177               }                                                               \
 178             else                                                              \
 179               {                                                               \
 180                 /* No JIS 0208 character.  */                                 \
 181                 found = ucs4_to_jisx0212 (ch, outptr + 1,                     \
 182                                           outend - outptr - 1);               \
 183                                                                               \
 184                 if (__builtin_expect (found, 1) == 0)                         \
 185                   {                                                           \
 186                     /* We ran out of space.  */                               \
 187                     result = __GCONV_FULL_OUTPUT;                             \
 188                     break;                                                    \
 189                   }                                                           \
 190                 else if (__builtin_expect (found, 0) != __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                     UNICODE_TAG_HANDLER (ch, 4);                              \
 200                                                                               \
 201                     /* Illegal character.  */                                 \
 202                     STANDARD_TO_LOOP_ERR_HANDLER (4);                         \
 203                   }                                                           \
 204               }                                                               \
 205           }                                                                   \
 206       }                                                                       \
 207                                                                               \
 208     inptr += 4;                                                               \
 209   }
 210 #define LOOP_NEED_FLAGS
 211 #include <iconv/loop.c>
 212
 213
 214 /* Now define the toplevel functions.  */
 215 #include <iconv/skeleton.c>