iconvdata/euc-cn.c

   1 /* Mapping tables for EUC-CN handling.
   2    Copyright (C) 1998, 1999, 2000, 2001 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 Lesser General Public
   8    License as published by the Free Software Foundation; either
   9    version 2.1 of the 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    Lesser General Public License for more details.
  15
  16    You should have received a copy of the GNU Lesser General Public
  17    License along with the GNU C Library; if not, write to the Free
  18    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  19    02111-1307 USA.  */
  20
  21 #include <dlfcn.h>
  22 #include <gb2312.h>
  23 #include <stdint.h>
  24
  25 /* Definitions used in the body of the `gconv' function.  */
  26 #define CHARSET_NAME            "EUC-CN//"
  27 #define FROM_LOOP               from_euc_cn
  28 #define TO_LOOP                 to_euc_cn
  29 #define DEFINE_INIT             1
  30 #define DEFINE_FINI             1
  31 #define MIN_NEEDED_FROM         1
  32 #define MAX_NEEDED_FROM         2
  33 #define MIN_NEEDED_TO           4
  34
  35
  36 /* First define the conversion function from EUC-CN 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       ++inptr;                                                                \
  47     else                                                                      \
  48       if ((__builtin_expect (ch <= 0xa0, 0) && ch != 0x8e && ch != 0x8f)      \
  49           || __builtin_expect (ch > 0xfe, 0))                                 \
  50         {                                                                     \
  51           /* This is illegal.  */                                             \
  52           if (! ignore_errors_p ())                                           \
  53             {                                                                 \
  54               result = __GCONV_ILLEGAL_INPUT;                                 \
  55               break;                                                          \
  56             }                                                                 \
  57                                                                               \
  58           ++inptr;                                                            \
  59           ++*irreversible;                                                    \
  60           continue;                                                           \
  61         }                                                                     \
  62       else                                                                    \
  63         {                                                                     \
  64           /* Two or more byte character.  First test whether the              \
  65              next character is also available.  */                            \
  66           const unsigned char *endp;                                          \
  67                                                                               \
  68           if (__builtin_expect (inptr + 1 >= inend, 0))                       \
  69             {                                                                 \
  70               /* The second character is not available.  Store                \
  71                  the intermediate result.  */                                 \
  72               result = __GCONV_INCOMPLETE_INPUT;                              \
  73               break;                                                          \
  74             }                                                                 \
  75                                                                               \
  76           ch = inptr[1];                                                      \
  77                                                                               \
  78           /* All second bytes of a multibyte character must be >= 0xa1. */    \
  79           if (__builtin_expect (ch < 0xa1, 0))                                \
  80             {                                                                 \
  81               if (! ignore_errors_p ())                                       \
  82                 {                                                             \
  83                   /* This is an illegal character.  */                        \
  84                   result = __GCONV_ILLEGAL_INPUT;                             \
  85                   break;                                                      \
  86                 }                                                             \
  87                                                                               \
  88               ++inptr;                                                        \
  89               ++*irreversible;                                                \
  90               continue;                                                       \
  91             }                                                                 \
  92                                                                               \
  93           /* This is code set 1: GB 2312-80.  */                              \
  94           endp = inptr;                                                       \
  95                                                                               \
  96           ch = gb2312_to_ucs4 (&endp, 2, 0x80);                               \
  97           if (__builtin_expect (ch, 0) == __UNKNOWN_10646_CHAR)               \
  98             {                                                                 \
  99               /* This is an illegal character.  */                            \
 100               if (! ignore_errors_p ())                                       \
 101                 {                                                             \
 102                   /* This is an illegal character.  */                        \
 103                   result = __GCONV_ILLEGAL_INPUT;                             \
 104                   break;                                                      \
 105                 }                                                             \
 106                                                                               \
 107               inptr += 2;                                                     \
 108               ++*irreversible;                                                \
 109               continue;                                                       \
 110             }                                                                 \
 111                                                                               \
 112           inptr += 2;                                                         \
 113         }                                                                     \
 114                                                                               \
 115     put32 (outptr, ch);                                                       \
 116     outptr += 4;                                                              \
 117   }
 118 #define LOOP_NEED_FLAGS
 119 #include <iconv/loop.c>
 120
 121
 122 /* Next, define the other direction.  */
 123 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 124 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 125 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 126 #define LOOPFCT                 TO_LOOP
 127 #define BODY \
 128   {                                                                           \
 129     uint32_t ch = get32 (inptr);                                              \
 130                                                                               \
 131     if (ch <= L'\x7f')                                                        \
 132       /* It's plain ASCII.  */                                                \
 133       *outptr++ = (unsigned char) ch;                                         \
 134     else                                                                      \
 135       {                                                                       \
 136         size_t found;                                                         \
 137                                                                               \
 138         found = ucs4_to_gb2312 (ch, outptr, outend - outptr);                 \
 139         if (__builtin_expect (found, 1) != 0)                                 \
 140           {                                                                   \
 141             if (__builtin_expect (found, 0) == __UNKNOWN_10646_CHAR)          \
 142               {                                                               \
 143                 UNICODE_TAG_HANDLER (ch, 4);                                  \
 144                                                                               \
 145                 /* Illegal character.  */                                     \
 146                 STANDARD_ERR_HANDLER (4);                                     \
 147               }                                                               \
 148                                                                               \
 149             /* It's a GB 2312 character, adjust it for EUC-CN.  */            \
 150             *outptr++ += 0x80;                                                \
 151             *outptr++ += 0x80;                                                \
 152           }                                                                   \
 153         else                                                                  \
 154           {                                                                   \
 155             /* We ran out of space.  */                                       \
 156             result = __GCONV_FULL_OUTPUT;                                     \
 157             break;                                                            \
 158           }                                                                   \
 159       }                                                                       \
 160     inptr += 4;                                                               \
 161   }
 162 #define LOOP_NEED_FLAGS
 163 #include <iconv/loop.c>
 164
 165
 166 /* Now define the toplevel functions.  */
 167 #include <iconv/skeleton.c>