iconvdata/iso-2022-kr.c

   1 /* Conversion module for ISO-2022-KR.
   2    Copyright (C) 1998, 1999, 2000 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 Library General Public License as
   8    published by the Free Software Foundation; either version 2 of the
   9    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    Library General Public License for more details.
  15
  16    You should have received a copy of the GNU Library General Public
  17    License along with the GNU C Library; see the file COPYING.LIB.  If not,
  18    write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  19    Boston, MA 02111-1307, USA.  */
  20
  21 #include <gconv.h>
  22 #include <stdint.h>
  23 #include <string.h>
  24 #include "ksc5601.h"
  25
  26 #include <assert.h>
  27
  28 /* This makes obvious what everybody knows: 0x1b is the Esc character.  */
  29 #define ESC     0x1b
  30
  31 /* The shift sequences for this charset (it does not use ESC).  */
  32 #define SI      0x0f
  33 #define SO      0x0e
  34
  35 /* Definitions used in the body of the `gconv' function.  */
  36 #define CHARSET_NAME            "ISO-2022-KR//"
  37 #define DEFINE_INIT             1
  38 #define DEFINE_FINI             1
  39 #define FROM_LOOP               from_iso2022kr_loop
  40 #define TO_LOOP                 to_iso2022kr_loop
  41 #define MIN_NEEDED_FROM         1
  42 #define MAX_NEEDED_FROM         3
  43 #define MIN_NEEDED_TO           4
  44 #define MAX_NEEDED_TO           4
  45 #define PREPARE_LOOP \
  46   int save_set;                                                               \
  47   int *setp = &data->__statep->count;                                         \
  48   if (!FROM_DIRECTION && !data->__internal_use                                \
  49       && data->__invocation_counter == 0)                                     \
  50     {                                                                         \
  51       /* Emit the designator sequence.  */                                    \
  52       if (outbuf + 4 > outend)                                                \
  53         return __GCONV_FULL_OUTPUT;                                           \
  54                                                                               \
  55       *outbuf++ = ESC;                                                        \
  56       *outbuf++ = '$';                                                        \
  57       *outbuf++ = ')';                                                        \
  58       *outbuf++ = 'C';                                                        \
  59     }
  60 #define EXTRA_LOOP_ARGS         , setp
  61
  62
  63 /* The COUNT element of the state keeps track of the currently selected
  64    character set.  The possible values are:  */
  65 enum
  66 {
  67   ASCII_set = 0,
  68   KSC5601_set
  69 };
  70
  71
  72 /* Since this is a stateful encoding we have to provide code which resets
  73    the output state to the initial state.  This has to be done during the
  74    flushing.  */
  75 #define EMIT_SHIFT_TO_INIT \
  76   if (data->__statep->count != ASCII_set)                                     \
  77     {                                                                         \
  78       if (FROM_DIRECTION)                                                     \
  79         /* It's easy, we don't have to emit anything, we just reset the       \
  80            state for the input.  */                                           \
  81         data->__statep->count = ASCII_set;                                    \
  82       else                                                                    \
  83         {                                                                     \
  84           unsigned char *outbuf = data->__outbuf;                             \
  85                                                                               \
  86           /* We are not in the initial state.  To switch back we have         \
  87              to emit `SI'.  */                                                \
  88           if (outbuf == data->__outbufend)                                    \
  89             /* We don't have enough room in the output buffer.  */            \
  90             status = __GCONV_FULL_OUTPUT;                                     \
  91           else                                                                \
  92             {                                                                 \
  93               /* Write out the shift sequence.  */                            \
  94               *outbuf++ = SI;                                                 \
  95               if (data->__is_last)                                            \
  96                 *written += 1;                                                \
  97               data->__outbuf = outbuf;                                        \
  98               data->__statep->count = ASCII_set;                              \
  99             }                                                                 \
 100         }                                                                     \
 101     }
 102
 103
 104 /* Since we might have to reset input pointer we must be able to save
 105    and retore the state.  */
 106 #define SAVE_RESET_STATE(Save) \
 107   if (Save)                                                                   \
 108     save_set = *setp;                                                         \
 109   else                                                                        \
 110     *setp = save_set
 111
 112
 113 /* First define the conversion function from ISO-2022-KR to UCS4.  */
 114 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
 115 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
 116 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
 117 #define LOOPFCT                 FROM_LOOP
 118 #define BODY \
 119   {                                                                           \
 120     uint32_t ch = *inptr;                                                     \
 121                                                                               \
 122     /* This is a 7bit character set, disallow all 8bit characters.  */        \
 123     if (ch > 0x7f)                                                            \
 124       {                                                                       \
 125         result = __GCONV_ILLEGAL_INPUT;                                       \
 126         break;                                                                \
 127       }                                                                       \
 128                                                                               \
 129     /* Recognize escape sequences.  */                                        \
 130     if (ch == ESC)                                                            \
 131       {                                                                       \
 132         /* We don't really have to handle escape sequences since all the      \
 133            switching is done using the SI and SO bytes.  But we have to       \
 134            recognize `Esc $ ) C' since this is a kind of flag for this        \
 135            encoding.  We simply ignore it.  */                                \
 136         if ((NEED_LENGTH_TEST && inptr + 1 > inend)                           \
 137             || (inptr[1] == '$'                                               \
 138                 && ((NEED_LENGTH_TEST && inptr + 2 > inend)                   \
 139                     || (inptr[2] == ')' && inptr + 3 > inend))))              \
 140                                                                               \
 141           {                                                                   \
 142             result = __GCONV_EMPTY_INPUT;                                     \
 143             break;                                                            \
 144           }                                                                   \
 145         if (inptr[1] == '$' && inptr[2] == ')' && inptr[3] == 'C')            \
 146           {                                                                   \
 147             /* Yeah, yeah, we know this is ISO 2022-KR.  */                   \
 148             inptr += 4;                                                       \
 149             continue;                                                         \
 150           }                                                                   \
 151       }                                                                       \
 152     else if (ch == SO)                                                        \
 153       {                                                                       \
 154         /* Switch to use KSC.  */                                             \
 155         ++inptr;                                                              \
 156         set = KSC5601_set;                                                    \
 157         continue;                                                             \
 158       }                                                                       \
 159     else if (ch == SI)                                                        \
 160       {                                                                       \
 161         /* Switch to use ASCII.  */                                           \
 162         ++inptr;                                                              \
 163         set = ASCII_set;                                                      \
 164         continue;                                                             \
 165       }                                                                       \
 166                                                                               \
 167     if (set == ASCII_set)                                                     \
 168       {                                                                       \
 169         /* Almost done, just advance the input pointer.  */                   \
 170         ++inptr;                                                              \
 171       }                                                                       \
 172     else                                                                      \
 173       {                                                                       \
 174         assert (set == KSC5601_set);                                          \
 175                                                                               \
 176         /* Use the KSC 5601 table.  */                                        \
 177         ch = ksc5601_to_ucs4 (&inptr,                                         \
 178                               NEED_LENGTH_TEST ? inend - inptr : 2, 0);       \
 179                                                                               \
 180         if (NEED_LENGTH_TEST && ch == 0)                                      \
 181           {                                                                   \
 182             result = __GCONV_EMPTY_INPUT;                                     \
 183             break;                                                            \
 184           }                                                                   \
 185         else if (ch == __UNKNOWN_10646_CHAR)                                  \
 186           {                                                                   \
 187             result = __GCONV_ILLEGAL_INPUT;                                   \
 188             break;                                                            \
 189           }                                                                   \
 190       }                                                                       \
 191                                                                               \
 192     put32 (outptr, ch);                                                       \
 193     outptr += 4;                                                              \
 194   }
 195 #define EXTRA_LOOP_DECLS        , int *setp
 196 #define INIT_PARAMS             int set = *setp
 197 #define UPDATE_PARAMS           *setp = set
 198 #include <iconv/loop.c>
 199
 200
 201 /* Next, define the other direction.  */
 202 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 203 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 204 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 205 #define LOOPFCT                 TO_LOOP
 206 #define BODY \
 207   {                                                                           \
 208     uint32_t ch;                                                              \
 209     size_t written = 0;                                                       \
 210                                                                               \
 211     ch = get32 (inptr);                                                       \
 212                                                                               \
 213     /* First see whether we can write the character using the currently       \
 214        selected character set.  */                                            \
 215     if (ch < 0x80)                                                            \
 216       {                                                                       \
 217         if (set != ASCII_set)                                                 \
 218           {                                                                   \
 219             *outptr++ = SI;                                                   \
 220             set = ASCII_set;                                                  \
 221             if (NEED_LENGTH_TEST && outptr == outend)                         \
 222               {                                                               \
 223                 result = __GCONV_FULL_OUTPUT;                                 \
 224                 break;                                                        \
 225               }                                                               \
 226           }                                                                   \
 227                                                                               \
 228         *outptr++ = ch;                                                       \
 229         written = 1;                                                          \
 230       }                                                                       \
 231     else                                                                      \
 232       {                                                                       \
 233         char buf[2];                                                          \
 234                                                                               \
 235         written = ucs4_to_ksc5601 (ch, buf, 2);                               \
 236                                                                               \
 237         if (written == __UNKNOWN_10646_CHAR)                                  \
 238           {                                                                   \
 239             /* Illegal character.  */                                         \
 240             result = __GCONV_ILLEGAL_INPUT;                                   \
 241             break;                                                            \
 242           }                                                                   \
 243         assert (written == 2);                                                \
 244                                                                               \
 245         /* We use KSC 5601.  */                                               \
 246         if (set != KSC5601_set)                                               \
 247           {                                                                   \
 248             *outptr++ = SO;                                                   \
 249             set = KSC5601_set;                                                \
 250           }                                                                   \
 251                                                                               \
 252         if (NEED_LENGTH_TEST && outptr + 2 > outend)                          \
 253           {                                                                   \
 254             result = __GCONV_FULL_OUTPUT;                                     \
 255             break;                                                            \
 256           }                                                                   \
 257                                                                               \
 258         *outptr++ = buf[0];                                                   \
 259         *outptr++ = buf[1];                                                   \
 260       }                                                                       \
 261                                                                               \
 262     /* Now that we wrote the output increment the input pointer.  */          \
 263     inptr += 4;                                                               \
 264   }
 265 #define EXTRA_LOOP_DECLS        , int *setp
 266 #define INIT_PARAMS             int set = *setp
 267 #define UPDATE_PARAMS           *setp = set
 268 #include <iconv/loop.c>
 269
 270
 271 /* Now define the toplevel functions.  */
 272 #include <iconv/skeleton.c>