iconvdata/euc-kr.c

   1 /* Mapping tables for EUC-KR handling.
   2    Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4    Contributed by Jungshik Shin <jshin@pantheon.yale.edu>
   5    and Ulrich Drepper <drepper@cygnus.com>, 1998.
   6
   7    The GNU C Library is free software; you can redistribute it and/or
   8    modify it under the terms of the GNU Library General Public License as
   9    published by the Free Software Foundation; either version 2 of the
  10    License, or (at your option) any later version.
  11
  12    The GNU C Library is distributed in the hope that it will be useful,
  13    but WITHOUT ANY WARRANTY; without even the implied warranty of
  14    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15    Library General Public License for more details.
  16
  17    You should have received a copy of the GNU Library General Public
  18    License along with the GNU C Library; see the file COPYING.LIB.  If not,
  19    write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  20    Boston, MA 02111-1307, USA.  */
  21
  22 #include <dlfcn.h>
  23 #include <stdint.h>
  24 #include <ksc5601.h>
  25
  26
  27 static inline void
  28 euckr_from_ucs4 (uint32_t ch, unsigned char *cp)
  29 {
  30   if (ch > 0x9f)
  31     {
  32       if (__builtin_expect (ch, 0) == 0x20a9)
  33         {
  34           /* Half-width Korean Currency WON sign.  There is no
  35              equivalent in EUC-KR.  Some mappings use \x5c because
  36              this is what some old Korean ASCII variants used but this
  37              is causing problems.  We map it to the FULL WIDTH WON SIGN.  */
  38           cp[0] = '\xa3';
  39           cp[1] = '\xdc';
  40         }
  41       else if (__builtin_expect (ucs4_to_ksc5601 (ch, cp, 2), 0)
  42           != __UNKNOWN_10646_CHAR)
  43         {
  44           cp[0] |= 0x80;
  45           cp[1] |= 0x80;
  46         }
  47       else
  48         cp[0] = '\0';
  49     }
  50   else
  51     {
  52       /* There is no mapping for U005c but we nevertheless map it to
  53          \x5c.  */
  54       cp[0] = (unsigned char) ch;
  55       cp[1] = '\0';
  56     }
  57 }
  58
  59
  60 /* Definitions used in the body of the `gconv' function.  */
  61 #define CHARSET_NAME            "EUC-KR//"
  62 #define FROM_LOOP               from_euc_kr
  63 #define TO_LOOP                 to_euc_kr
  64 #define DEFINE_INIT             1
  65 #define DEFINE_FINI             1
  66 #define MIN_NEEDED_FROM         1
  67 #define MAX_NEEDED_FROM         2
  68 #define MIN_NEEDED_TO           4
  69
  70
  71 /* First define the conversion function from EUC-KR to UCS4.  */
  72 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  73 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  74 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  75 #define LOOPFCT                 FROM_LOOP
  76 #define BODY \
  77   {                                                                           \
  78     uint32_t ch = *inptr;                                                     \
  79                                                                               \
  80     if (ch <= 0x9f)                                                           \
  81       ++inptr;                                                                \
  82     /* 0xfe(->0x7e : row 94) and 0xc9(->0x59 : row 41) are                    \
  83        user-defined areas.  */                                                \
  84     else if (__builtin_expect (ch, 0xa1) == 0xa0                              \
  85              || __builtin_expect (ch, 0xa1) > 0xfe                            \
  86              || __builtin_expect (ch, 0xa1) == 0xc9)                          \
  87       {                                                                       \
  88         /* This is illegal.  */                                               \
  89         if (! ignore_errors_p ())                                             \
  90           {                                                                   \
  91             result = __GCONV_ILLEGAL_INPUT;                                   \
  92             break;                                                            \
  93           }                                                                   \
  94                                                                               \
  95         ++inptr;                                                              \
  96         ++*irreversible;                                                      \
  97         continue;                                                             \
  98       }                                                                       \
  99     else                                                                      \
 100       {                                                                       \
 101         /* Two-byte character.  First test whether the next character         \
 102            is also available.  */                                             \
 103         ch = ksc5601_to_ucs4 (&inptr, inend - inptr, 0x80);                   \
 104         if (__builtin_expect (ch, 1) == 0)                                    \
 105           {                                                                   \
 106             /* The second character is not available.  */                     \
 107             result = __GCONV_INCOMPLETE_INPUT;                                \
 108             break;                                                            \
 109           }                                                                   \
 110         if (__builtin_expect (ch, 0) == __UNKNOWN_10646_CHAR)                 \
 111           {                                                                   \
 112             /* This is an illegal character.  */                              \
 113             if (! ignore_errors_p ())                                         \
 114               {                                                               \
 115                 /* This is an illegal character.  */                          \
 116                 result = __GCONV_ILLEGAL_INPUT;                               \
 117                 break;                                                        \
 118               }                                                               \
 119                                                                               \
 120             inptr += 2;                                                       \
 121             ++*irreversible;                                                  \
 122             continue;                                                         \
 123           }                                                                   \
 124       }                                                                       \
 125                                                                               \
 126     put32 (outptr, ch);                                                       \
 127     outptr += 4;                                                              \
 128   }
 129 #define LOOP_NEED_FLAGS
 130 #include <iconv/loop.c>
 131
 132
 133 /* Next, define the other direction.  */
 134 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 135 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 136 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 137 #define LOOPFCT                 TO_LOOP
 138 #define BODY \
 139   {                                                                           \
 140     uint32_t ch = get32 (inptr);                                              \
 141     unsigned char cp[2];                                                      \
 142                                                                               \
 143     /* Decomposing Hangul syllables not available in KS C 5601 into           \
 144        Jamos should be considered either here or in euckr_from_ucs4() */      \
 145     euckr_from_ucs4 (ch, cp) ;                                                \
 146                                                                               \
 147     if (__builtin_expect (cp[0], '\1') == '\0' && ch != 0)                    \
 148       {                                                                       \
 149         /* Illegal character.  */                                             \
 150         STANDARD_ERR_HANDLER (4);                                             \
 151       }                                                                       \
 152                                                                               \
 153     *outptr++ = cp[0];                                                        \
 154     /* Now test for a possible second byte and write this if possible.  */    \
 155     if (cp[1] != '\0')                                                        \
 156       {                                                                       \
 157         if (__builtin_expect (outptr >= outend, 0))                           \
 158           {                                                                   \
 159             /* The result does not fit into the buffer.  */                   \
 160             --outptr;                                                         \
 161             result = __GCONV_FULL_OUTPUT;                                     \
 162             break;                                                            \
 163           }                                                                   \
 164         *outptr++ = cp[1];                                                    \
 165       }                                                                       \
 166                                                                               \
 167     inptr += 4;                                                               \
 168   }
 169 #define LOOP_NEED_FLAGS
 170 #include <iconv/loop.c>
 171
 172
 173 /* Now define the toplevel functions.  */
 174 #include <iconv/skeleton.c>