iconvdata/euc-cn.c

   1 /* Mapping tables for EUC-CN handling.
   2    Copyright (C) 1998-2019 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, see
  18    <https://www.gnu.org/licenses/>.  */
  19
  20 #include <dlfcn.h>
  21 #include <gb2312.h>
  22 #include <stdint.h>
  23
  24 /* Definitions used in the body of the `gconv' function.  */
  25 #define CHARSET_NAME            "EUC-CN//"
  26 #define FROM_LOOP               from_euc_cn
  27 #define TO_LOOP                 to_euc_cn
  28 #define DEFINE_INIT             1
  29 #define DEFINE_FINI             1
  30 #define MIN_NEEDED_FROM         1
  31 #define MAX_NEEDED_FROM         2
  32 #define MIN_NEEDED_TO           4
  33 #define ONE_DIRECTION           0
  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           STANDARD_FROM_LOOP_ERR_HANDLER (1);                                 \
  53         }                                                                     \
  54       else                                                                    \
  55         {                                                                     \
  56           /* Two or more byte character.  First test whether the              \
  57              next byte is also available.  */                                 \
  58           const unsigned char *endp;                                          \
  59                                                                               \
  60           if (__glibc_unlikely (inptr + 1 >= inend))                          \
  61             {                                                                 \
  62               /* The second character is not available.  Store                \
  63                  the intermediate result.  */                                 \
  64               result = __GCONV_INCOMPLETE_INPUT;                              \
  65               break;                                                          \
  66             }                                                                 \
  67                                                                               \
  68           ch = inptr[1];                                                      \
  69                                                                               \
  70           /* All second bytes of a multibyte character must be >= 0xa1. */    \
  71           if (__glibc_unlikely (ch < 0xa1))                                   \
  72             STANDARD_FROM_LOOP_ERR_HANDLER (1);                               \
  73                                                                               \
  74           /* This is code set 1: GB 2312-80.  */                              \
  75           endp = inptr;                                                       \
  76                                                                               \
  77           ch = gb2312_to_ucs4 (&endp, 2, 0x80);                               \
  78           if (__glibc_unlikely (ch == __UNKNOWN_10646_CHAR))                  \
  79             {                                                                 \
  80               /* This is an illegal character.  */                            \
  81               STANDARD_FROM_LOOP_ERR_HANDLER (2);                             \
  82             }                                                                 \
  83                                                                               \
  84           inptr += 2;                                                         \
  85         }                                                                     \
  86                                                                               \
  87     put32 (outptr, ch);                                                       \
  88     outptr += 4;                                                              \
  89   }
  90 #define LOOP_NEED_FLAGS
  91 #define ONEBYTE_BODY \
  92   {                                                                           \
  93     if (c < 0x80)                                                             \
  94       return c;                                                               \
  95     else                                                                      \
  96       return WEOF;                                                            \
  97   }
  98 #include <iconv/loop.c>
  99
 100
 101 /* Next, define the other direction.  */
 102 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 103 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 104 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 105 #define LOOPFCT                 TO_LOOP
 106 #define BODY \
 107   {                                                                           \
 108     uint32_t ch = get32 (inptr);                                              \
 109                                                                               \
 110     if (ch <= L'\x7f')                                                        \
 111       /* It's plain ASCII.  */                                                \
 112       *outptr++ = (unsigned char) ch;                                         \
 113     else                                                                      \
 114       {                                                                       \
 115         size_t found;                                                         \
 116                                                                               \
 117         found = ucs4_to_gb2312 (ch, outptr, outend - outptr);                 \
 118         if (__builtin_expect (found, 1) != 0)                                 \
 119           {                                                                   \
 120             if (__builtin_expect (found, 0) == __UNKNOWN_10646_CHAR)          \
 121               {                                                               \
 122                 UNICODE_TAG_HANDLER (ch, 4);                                  \
 123                                                                               \
 124                 /* Illegal character.  */                                     \
 125                 STANDARD_TO_LOOP_ERR_HANDLER (4);                             \
 126               }                                                               \
 127                                                                               \
 128             /* It's a GB 2312 character, adjust it for EUC-CN.  */            \
 129             *outptr++ += 0x80;                                                \
 130             *outptr++ += 0x80;                                                \
 131           }                                                                   \
 132         else                                                                  \
 133           {                                                                   \
 134             /* We ran out of space.  */                                       \
 135             result = __GCONV_FULL_OUTPUT;                                     \
 136             break;                                                            \
 137           }                                                                   \
 138       }                                                                       \
 139     inptr += 4;                                                               \
 140   }
 141 #define LOOP_NEED_FLAGS
 142 #include <iconv/loop.c>
 143
 144
 145 /* Now define the toplevel functions.  */
 146 #include <iconv/skeleton.c>