iconvdata/euc-kr.c

   1 /* Mapping tables for EUC-KR 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 <ksc5601.h>
  22
  23
  24 static inline void
  25 __attribute ((always_inline))
  26 euckr_from_ucs4 (uint32_t ch, unsigned char *cp)
  27 {
  28   if (ch > 0x9f)
  29     {
  30       if (__builtin_expect (ch, 0) == 0x20a9)
  31         {
  32           /* Half-width Korean Currency WON sign.  There is no
  33              equivalent in EUC-KR.  Some mappings use \x5c because
  34              this is what some old Korean ASCII variants used but this
  35              is causing problems.  We map it to the FULL WIDTH WON SIGN.  */
  36           cp[0] = '\xa3';
  37           cp[1] = '\xdc';
  38         }
  39       else if (__builtin_expect (ucs4_to_ksc5601 (ch, cp, 2), 0)
  40           != __UNKNOWN_10646_CHAR)
  41         {
  42           cp[0] |= 0x80;
  43           cp[1] |= 0x80;
  44         }
  45       else
  46         cp[0] = cp[1] = '\0';
  47     }
  48   else
  49     {
  50       /* There is no mapping for U005c but we nevertheless map it to
  51          \x5c.  */
  52       cp[0] = (unsigned char) ch;
  53       cp[1] = '\0';
  54     }
  55 }
  56
  57
  58 /* Definitions used in the body of the `gconv' function.  */
  59 #define CHARSET_NAME            "EUC-KR//"
  60 #define FROM_LOOP               from_euc_kr
  61 #define TO_LOOP                 to_euc_kr
  62 #define DEFINE_INIT             1
  63 #define DEFINE_FINI             1
  64 #define MIN_NEEDED_FROM         1
  65 #define MAX_NEEDED_FROM         2
  66 #define MIN_NEEDED_TO           4
  67 #define ONE_DIRECTION           0
  68
  69
  70 /* First define the conversion function from EUC-KR to UCS4.  */
  71 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  72 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  73 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  74 #define LOOPFCT                 FROM_LOOP
  75 #define BODY \
  76   {                                                                           \
  77     uint32_t ch = *inptr;                                                     \
  78                                                                               \
  79     if (ch <= 0x9f)                                                           \
  80       ++inptr;                                                                \
  81     else if (__glibc_unlikely (ch == 0xa0))                                   \
  82       {                                                                       \
  83         /* This is illegal.  */                                               \
  84         STANDARD_FROM_LOOP_ERR_HANDLER (1);                                   \
  85       }                                                                       \
  86     else                                                                      \
  87       {                                                                       \
  88         /* Two-byte character.  First test whether the next byte              \
  89            is also available.  */                                             \
  90         ch = ksc5601_to_ucs4 (&inptr, inend - inptr, 0x80);                   \
  91         if (__glibc_unlikely (ch == 0))                                       \
  92           {                                                                   \
  93             /* The second byte is not available.  */                          \
  94             result = __GCONV_INCOMPLETE_INPUT;                                \
  95             break;                                                            \
  96           }                                                                   \
  97         if (__glibc_unlikely (ch == __UNKNOWN_10646_CHAR))                    \
  98           /* This is an illegal character.  */                                \
  99           STANDARD_FROM_LOOP_ERR_HANDLER (2);                                 \
 100       }                                                                       \
 101                                                                               \
 102     put32 (outptr, ch);                                                       \
 103     outptr += 4;                                                              \
 104   }
 105 #define LOOP_NEED_FLAGS
 106 #define ONEBYTE_BODY \
 107   {                                                                           \
 108     if (c <= 0x9f)                                                            \
 109       return c;                                                               \
 110     else                                                                      \
 111       return WEOF;                                                            \
 112   }
 113 #include <iconv/loop.c>
 114
 115
 116 /* Next, define the other direction.  */
 117 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 118 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 119 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 120 #define LOOPFCT                 TO_LOOP
 121 #define BODY \
 122   {                                                                           \
 123     uint32_t ch = get32 (inptr);                                              \
 124     unsigned char cp[2];                                                      \
 125                                                                               \
 126     /* Decomposing Hangul syllables not available in KS C 5601 into           \
 127        Jamos should be considered either here or in euckr_from_ucs4() */      \
 128     euckr_from_ucs4 (ch, cp);                                                 \
 129                                                                               \
 130     if (__builtin_expect (cp[0], '\1') == '\0' && ch != 0)                    \
 131       {                                                                       \
 132         UNICODE_TAG_HANDLER (ch, 4);                                          \
 133                                                                               \
 134         /* Illegal character.  */                                             \
 135         STANDARD_TO_LOOP_ERR_HANDLER (4);                                     \
 136       }                                                                       \
 137                                                                               \
 138     *outptr++ = cp[0];                                                        \
 139     /* Now test for a possible second byte and write this if possible.  */    \
 140     if (cp[1] != '\0')                                                        \
 141       {                                                                       \
 142         if (__glibc_unlikely (outptr >= outend))                              \
 143           {                                                                   \
 144             /* The result does not fit into the buffer.  */                   \
 145             --outptr;                                                         \
 146             result = __GCONV_FULL_OUTPUT;                                     \
 147             break;                                                            \
 148           }                                                                   \
 149         *outptr++ = cp[1];                                                    \
 150       }                                                                       \
 151                                                                               \
 152     inptr += 4;                                                               \
 153   }
 154 #define LOOP_NEED_FLAGS
 155 #include <iconv/loop.c>
 156
 157
 158 /* Now define the toplevel functions.  */
 159 #include <iconv/skeleton.c>