iconvdata/iso-2022-kr.c

   1 /* Conversion module for ISO-2022-KR.
   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 <gconv.h>
  21 #include <stdint.h>
  22 #include <string.h>
  23 #include "ksc5601.h"
  24
  25 #include <assert.h>
  26
  27 /* This makes obvious what everybody knows: 0x1b is the Esc character.  */
  28 #define ESC     0x1b
  29
  30 /* The shift sequences for this charset (it does not use ESC).  */
  31 #define SI      0x0f
  32 #define SO      0x0e
  33
  34 /* Definitions used in the body of the `gconv' function.  */
  35 #define CHARSET_NAME            "ISO-2022-KR//"
  36 #define DEFINE_INIT             1
  37 #define DEFINE_FINI             1
  38 #define FROM_LOOP               from_iso2022kr_loop
  39 #define TO_LOOP                 to_iso2022kr_loop
  40 #define MIN_NEEDED_FROM         1
  41 #define MAX_NEEDED_FROM         4
  42 #define MIN_NEEDED_TO           4
  43 #define MAX_NEEDED_TO           4
  44 #define ONE_DIRECTION           0
  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           /* We are not in the initial state.  To switch back we have         \
  88              to emit `SI'.  */                                                \
  89           if (__glibc_unlikely (outbuf == outend))                            \
  90             /* We don't have enough room in the output buffer.  */            \
  91             status = __GCONV_FULL_OUTPUT;                                     \
  92           else                                                                \
  93             {                                                                 \
  94               /* Write out the shift sequence.  */                            \
  95               *outbuf++ = SI;                                                 \
  96               data->__statep->__count = ASCII_set;                            \
  97             }                                                                 \
  98         }                                                                     \
  99     }
 100
 101
 102 /* Since we might have to reset input pointer we must be able to save
 103    and retore the state.  */
 104 #define SAVE_RESET_STATE(Save) \
 105   if (Save)                                                                   \
 106     save_set = *setp;                                                         \
 107   else                                                                        \
 108     *setp = save_set
 109
 110
 111 /* First define the conversion function from ISO-2022-KR to UCS4.  */
 112 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
 113 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
 114 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
 115 #define LOOPFCT                 FROM_LOOP
 116 #define BODY \
 117   {                                                                           \
 118     uint32_t ch = *inptr;                                                     \
 119                                                                               \
 120     /* This is a 7bit character set, disallow all 8bit characters.  */        \
 121     if (__glibc_unlikely (ch > 0x7f))                                         \
 122       STANDARD_FROM_LOOP_ERR_HANDLER (1);                                     \
 123                                                                               \
 124     /* Recognize escape sequences.  */                                        \
 125     if (__builtin_expect (ch, 0) == ESC)                                      \
 126       {                                                                       \
 127         /* We don't really have to handle escape sequences since all the      \
 128            switching is done using the SI and SO bytes.  But we have to       \
 129            recognize `Esc $ ) C' since this is a kind of flag for this        \
 130            encoding.  We simply ignore it.  */                                \
 131         if (__builtin_expect (inptr + 2 > inend, 0)                           \
 132             || (inptr[1] == '$'                                               \
 133                 && (__builtin_expect (inptr + 3 > inend, 0)                   \
 134                     || (inptr[2] == ')'                                       \
 135                         && __builtin_expect (inptr + 4 > inend, 0)))))        \
 136           {                                                                   \
 137             result = __GCONV_INCOMPLETE_INPUT;                                \
 138             break;                                                            \
 139           }                                                                   \
 140         if (inptr[1] == '$' && inptr[2] == ')' && inptr[3] == 'C')            \
 141           {                                                                   \
 142             /* Yeah, yeah, we know this is ISO 2022-KR.  */                   \
 143             inptr += 4;                                                       \
 144             continue;                                                         \
 145           }                                                                   \
 146       }                                                                       \
 147     else if (__builtin_expect (ch, 0) == SO)                                  \
 148       {                                                                       \
 149         /* Switch to use KSC.  */                                             \
 150         ++inptr;                                                              \
 151         set = KSC5601_set;                                                    \
 152         continue;                                                             \
 153       }                                                                       \
 154     else if (__builtin_expect (ch, 0) == SI)                                  \
 155       {                                                                       \
 156         /* Switch to use ASCII.  */                                           \
 157         ++inptr;                                                              \
 158         set = ASCII_set;                                                      \
 159         continue;                                                             \
 160       }                                                                       \
 161                                                                               \
 162     if (set == ASCII_set)                                                     \
 163       {                                                                       \
 164         /* Almost done, just advance the input pointer.  */                   \
 165         ++inptr;                                                              \
 166       }                                                                       \
 167     else                                                                      \
 168       {                                                                       \
 169         assert (set == KSC5601_set);                                          \
 170                                                                               \
 171         /* Use the KSC 5601 table.  */                                        \
 172         ch = ksc5601_to_ucs4 (&inptr, inend - inptr, 0);                      \
 173                                                                               \
 174         if (__glibc_unlikely (ch == 0))                                       \
 175           {                                                                   \
 176             result = __GCONV_INCOMPLETE_INPUT;                                \
 177             break;                                                            \
 178           }                                                                   \
 179         else if (__glibc_unlikely (ch == __UNKNOWN_10646_CHAR))               \
 180           {                                                                   \
 181             STANDARD_FROM_LOOP_ERR_HANDLER (1);                               \
 182           }                                                                   \
 183       }                                                                       \
 184                                                                               \
 185     put32 (outptr, ch);                                                       \
 186     outptr += 4;                                                              \
 187   }
 188 #define LOOP_NEED_FLAGS
 189 #define EXTRA_LOOP_DECLS        , int *setp
 190 #define INIT_PARAMS             int set = *setp
 191 #define UPDATE_PARAMS           *setp = set
 192 #include <iconv/loop.c>
 193
 194
 195 /* Next, define the other direction.  */
 196 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 197 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 198 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 199 #define LOOPFCT                 TO_LOOP
 200 #define BODY \
 201   {                                                                           \
 202     uint32_t ch = get32 (inptr);                                              \
 203                                                                               \
 204     /* First see whether we can write the character using the currently       \
 205        selected character set.  */                                            \
 206     if (ch < 0x80)                                                            \
 207       {                                                                       \
 208         if (set != ASCII_set)                                                 \
 209           {                                                                   \
 210             *outptr++ = SI;                                                   \
 211             set = ASCII_set;                                                  \
 212             if (__glibc_unlikely (outptr == outend))                          \
 213               {                                                               \
 214                 result = __GCONV_FULL_OUTPUT;                                 \
 215                 break;                                                        \
 216               }                                                               \
 217           }                                                                   \
 218                                                                               \
 219         *outptr++ = ch;                                                       \
 220       }                                                                       \
 221     else                                                                      \
 222       {                                                                       \
 223         unsigned char buf[2];                                                 \
 224         /* Fake initialization to keep gcc quiet.  */                         \
 225         asm ("" : "=m" (buf));                                                \
 226                                                                               \
 227         size_t written = ucs4_to_ksc5601 (ch, buf, 2);                        \
 228         if (__builtin_expect (written, 0) == __UNKNOWN_10646_CHAR)            \
 229           {                                                                   \
 230             UNICODE_TAG_HANDLER (ch, 4);                                      \
 231                                                                               \
 232             /* Illegal character.  */                                         \
 233             STANDARD_TO_LOOP_ERR_HANDLER (4);                                 \
 234           }                                                                   \
 235         else                                                                  \
 236           {                                                                   \
 237             assert (written == 2);                                            \
 238                                                                               \
 239             /* We use KSC 5601.  */                                           \
 240             if (set != KSC5601_set)                                           \
 241               {                                                               \
 242                 *outptr++ = SO;                                               \
 243                 set = KSC5601_set;                                            \
 244               }                                                               \
 245                                                                               \
 246             if (__glibc_unlikely (outptr + 2 > outend))                       \
 247               {                                                               \
 248                 result = __GCONV_FULL_OUTPUT;                                 \
 249                 break;                                                        \
 250               }                                                               \
 251                                                                               \
 252             *outptr++ = buf[0];                                               \
 253             *outptr++ = buf[1];                                               \
 254           }                                                                   \
 255       }                                                                       \
 256                                                                               \
 257     /* Now that we wrote the output increment the input pointer.  */          \
 258     inptr += 4;                                                               \
 259   }
 260 #define LOOP_NEED_FLAGS
 261 #define EXTRA_LOOP_DECLS        , int *setp
 262 #define INIT_PARAMS             int set = *setp
 263 #define REINIT_PARAMS           set = *setp
 264 #define UPDATE_PARAMS           *setp = set
 265 #include <iconv/loop.c>
 266
 267
 268 /* Now define the toplevel functions.  */
 269 #include <iconv/skeleton.c>