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 = 8
  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         {                                                                     \
  80           /* It's easy, we don't have to emit anything, we just reset the     \
  81              state for the input.  */                                         \
  82           data->__statep->__count &= 7;                                       \
  83           data->__statep->__count |= ASCII_set;                               \
  84         }                                                                     \
  85       else                                                                    \
  86         {                                                                     \
  87           unsigned char *outbuf = data->__outbuf;                             \
  88                                                                               \
  89           /* We are not in the initial state.  To switch back we have         \
  90              to emit `SI'.  */                                                \
  91           if (outbuf == data->__outbufend)                                    \
  92             /* We don't have enough room in the output buffer.  */            \
  93             status = __GCONV_FULL_OUTPUT;                                     \
  94           else                                                                \
  95             {                                                                 \
  96               /* Write out the shift sequence.  */                            \
  97               *outbuf++ = SI;                                                 \
  98               if (data->__is_last)                                            \
  99                 *written += 1;                                                \
 100               data->__outbuf = outbuf;                                        \
 101               data->__statep->__count = ASCII_set;                            \
 102             }                                                                 \
 103         }                                                                     \
 104     }
 105
 106
 107 /* Since we might have to reset input pointer we must be able to save
 108    and retore the state.  */
 109 #define SAVE_RESET_STATE(Save) \
 110   if (Save)                                                                   \
 111     save_set = *setp;                                                         \
 112   else                                                                        \
 113     *setp = save_set
 114
 115
 116 /* First define the conversion function from ISO-2022-KR to UCS4.  */
 117 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
 118 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
 119 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
 120 #define LOOPFCT                 FROM_LOOP
 121 #define BODY \
 122   {                                                                           \
 123     uint32_t ch = *inptr;                                                     \
 124                                                                               \
 125     /* This is a 7bit character set, disallow all 8bit characters.  */        \
 126     if (ch > 0x7f)                                                            \
 127       {                                                                       \
 128         result = __GCONV_ILLEGAL_INPUT;                                       \
 129         break;                                                                \
 130       }                                                                       \
 131                                                                               \
 132     /* Recognize escape sequences.  */                                        \
 133     if (ch == ESC)                                                            \
 134       {                                                                       \
 135         /* We don't really have to handle escape sequences since all the      \
 136            switching is done using the SI and SO bytes.  But we have to       \
 137            recognize `Esc $ ) C' since this is a kind of flag for this        \
 138            encoding.  We simply ignore it.  */                                \
 139         if ((NEED_LENGTH_TEST && inptr + 1 > inend)                           \
 140             || (inptr[1] == '$'                                               \
 141                 && ((NEED_LENGTH_TEST && inptr + 2 > inend)                   \
 142                     || (inptr[2] == ')' && inptr + 3 > inend))))              \
 143                                                                               \
 144           {                                                                   \
 145             result = __GCONV_EMPTY_INPUT;                                     \
 146             break;                                                            \
 147           }                                                                   \
 148         if (inptr[1] == '$' && inptr[2] == ')' && inptr[3] == 'C')            \
 149           {                                                                   \
 150             /* Yeah, yeah, we know this is ISO 2022-KR.  */                   \
 151             inptr += 4;                                                       \
 152             continue;                                                         \
 153           }                                                                   \
 154       }                                                                       \
 155     else if (ch == SO)                                                        \
 156       {                                                                       \
 157         /* Switch to use KSC.  */                                             \
 158         ++inptr;                                                              \
 159         set = KSC5601_set;                                                    \
 160         continue;                                                             \
 161       }                                                                       \
 162     else if (ch == SI)                                                        \
 163       {                                                                       \
 164         /* Switch to use ASCII.  */                                           \
 165         ++inptr;                                                              \
 166         set = ASCII_set;                                                      \
 167         continue;                                                             \
 168       }                                                                       \
 169                                                                               \
 170     if (set == ASCII_set)                                                     \
 171       {                                                                       \
 172         /* Almost done, just advance the input pointer.  */                   \
 173         ++inptr;                                                              \
 174       }                                                                       \
 175     else                                                                      \
 176       {                                                                       \
 177         assert (set == KSC5601_set);                                          \
 178                                                                               \
 179         /* Use the KSC 5601 table.  */                                        \
 180         ch = ksc5601_to_ucs4 (&inptr,                                         \
 181                               NEED_LENGTH_TEST ? inend - inptr : 2, 0);       \
 182                                                                               \
 183         if (NEED_LENGTH_TEST && ch == 0)                                      \
 184           {                                                                   \
 185             result = __GCONV_EMPTY_INPUT;                                     \
 186             break;                                                            \
 187           }                                                                   \
 188         else if (ch == __UNKNOWN_10646_CHAR)                                  \
 189           {                                                                   \
 190             result = __GCONV_ILLEGAL_INPUT;                                   \
 191             break;                                                            \
 192           }                                                                   \
 193       }                                                                       \
 194                                                                               \
 195     put32 (outptr, ch);                                                       \
 196     outptr += 4;                                                              \
 197   }
 198 #define EXTRA_LOOP_DECLS        , int *setp
 199 #define INIT_PARAMS             int set = *setp
 200 #define UPDATE_PARAMS           *setp = set
 201 #include <iconv/loop.c>
 202
 203
 204 /* Next, define the other direction.  */
 205 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 206 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 207 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 208 #define LOOPFCT                 TO_LOOP
 209 #define BODY \
 210   {                                                                           \
 211     uint32_t ch;                                                              \
 212     size_t written = 0;                                                       \
 213                                                                               \
 214     ch = get32 (inptr);                                                       \
 215                                                                               \
 216     /* First see whether we can write the character using the currently       \
 217        selected character set.  */                                            \
 218     if (ch < 0x80)                                                            \
 219       {                                                                       \
 220         if (set != ASCII_set)                                                 \
 221           {                                                                   \
 222             *outptr++ = SI;                                                   \
 223             set = ASCII_set;                                                  \
 224             if (NEED_LENGTH_TEST && outptr == outend)                         \
 225               {                                                               \
 226                 result = __GCONV_FULL_OUTPUT;                                 \
 227                 break;                                                        \
 228               }                                                               \
 229           }                                                                   \
 230                                                                               \
 231         *outptr++ = ch;                                                       \
 232         written = 1;                                                          \
 233       }                                                                       \
 234     else                                                                      \
 235       {                                                                       \
 236         char buf[2];                                                          \
 237                                                                               \
 238         written = ucs4_to_ksc5601 (ch, buf, 2);                               \
 239                                                                               \
 240         if (written == __UNKNOWN_10646_CHAR)                                  \
 241           {                                                                   \
 242             /* Illegal character.  */                                         \
 243             result = __GCONV_ILLEGAL_INPUT;                                   \
 244             break;                                                            \
 245           }                                                                   \
 246         assert (written == 2);                                                \
 247                                                                               \
 248         /* We use KSC 5601.  */                                               \
 249         if (set != KSC5601_set)                                               \
 250           {                                                                   \
 251             *outptr++ = SO;                                                   \
 252             set = KSC5601_set;                                                \
 253           }                                                                   \
 254                                                                               \
 255         if (NEED_LENGTH_TEST && outptr + 2 > outend)                          \
 256           {                                                                   \
 257             result = __GCONV_FULL_OUTPUT;                                     \
 258             break;                                                            \
 259           }                                                                   \
 260                                                                               \
 261         *outptr++ = buf[0];                                                   \
 262         *outptr++ = buf[1];                                                   \
 263       }                                                                       \
 264                                                                               \
 265     /* Now that we wrote the output increment the input pointer.  */          \
 266     inptr += 4;                                                               \
 267   }
 268 #define EXTRA_LOOP_DECLS        , int *setp
 269 #define INIT_PARAMS             int set = *setp
 270 #define UPDATE_PARAMS           *setp = set
 271 #include <iconv/loop.c>
 272
 273
 274 /* Now define the toplevel functions.  */
 275 #include <iconv/skeleton.c>