iconvdata/ibm937.c

   1 /* Conversion to and from IBM937.
   2    Copyright (C) 2000, 2001 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4    Contributed by Masahide Washizawa <washi@yamato.ibm.co.jp>, 2000.
   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 <dlfcn.h>
  22 #include <stdint.h>
  23 #include <wchar.h>
  24 #include <byteswap.h>
  25 #include "ibm937.h"
  26
  27 /* The shift sequences for this charset (it does not use ESC).  */
  28 #define SI              0x0F  /* Shift In, host code to turn DBCS off.  */
  29 #define SO              0x0E  /* Shift Out, host code to turn DBCS on.  */
  30
  31 /* Definitions used in the body of the `gconv' function.  */
  32 #define CHARSET_NAME    "IBM937//"
  33 #define FROM_LOOP       from_ibm937
  34 #define TO_LOOP         to_ibm937
  35 #define MIN_NEEDED_FROM 1
  36 #define MAX_NEEDED_FROM 2
  37 #define MIN_NEEDED_TO   4
  38 #define MAX_NEEDED_TO   4
  39 #define PREPARE_LOOP \
  40   int save_curcs;                                                             \
  41   int *curcsp = &data->__statep->__count;
  42 #define EXTRA_LOOP_ARGS         , curcsp
  43
  44 /* Definitions of initialization and destructor function.  */
  45 #define DEFINE_INIT     1
  46 #define DEFINE_FINI     1
  47
  48
  49 /* Since this is a stateful encoding we have to provide code which resets
  50    the output state to the initial state.  This has to be done during the
  51    flushing.  */
  52 #define EMIT_SHIFT_TO_INIT \
  53   if ((data->__statep->__count & ~7) != sb)                                   \
  54     {                                                                         \
  55       if (FROM_DIRECTION)                                                     \
  56         data->__statep->__count &= 7;                                         \
  57       else                                                                    \
  58         {                                                                     \
  59           unsigned char *outbuf = data->__outbuf;                             \
  60                                                                               \
  61           /* We are not in the initial state.  To switch back we have         \
  62              to emit `SI'.  */                                                \
  63           if (__builtin_expect (outbuf >= data->__outbufend, 0))              \
  64             /* We don't have enough room in the output buffer.  */            \
  65             status = __GCONV_FULL_OUTPUT;                                     \
  66           else                                                                \
  67             {                                                                 \
  68               /* Write out the shift sequence.  */                            \
  69               *outbuf++ = SI;                                                 \
  70               data->__outbuf = outbuf;                                        \
  71               data->__statep->__count &= 7;                                   \
  72             }                                                                 \
  73         }                                                                     \
  74     }
  75
  76
  77 /* Since we might have to reset input pointer we must be able to save
  78    and retore the state.  */
  79 #define SAVE_RESET_STATE(Save) \
  80   if (Save)                                                                   \
  81     save_curcs = *curcsp;                                                     \
  82   else                                                                        \
  83     *curcsp = save_curcs
  84
  85
  86 /* Current codeset type.  */
  87 enum
  88 {
  89   sb = 0,
  90   db = 64
  91 };
  92
  93 /* First, define the conversion function from IBM-937 to UCS4.  */
  94 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  95 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  96 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  97 #define LOOPFCT                 FROM_LOOP
  98 #define BODY \
  99   {                                                                           \
 100     uint32_t ch = *inptr;                                                     \
 101                                                                               \
 102     if (__builtin_expect (ch, 0) == SO)                                       \
 103       {                                                                       \
 104         /* Shift OUT, change to DBCS converter.  */                           \
 105         if (curcs == db)                                                      \
 106           {                                                                   \
 107             result = __GCONV_ILLEGAL_INPUT;                                   \
 108             break;                                                            \
 109           }                                                                   \
 110         curcs = db;                                                           \
 111         ++inptr;                                                              \
 112         continue;                                                             \
 113       }                                                                       \
 114     else if (__builtin_expect (ch, 0) == SI)                                  \
 115       {                                                                       \
 116         /* Shift IN, change to SBCS converter.  */                            \
 117         if (curcs == sb)                                                      \
 118           {                                                                   \
 119             result = __GCONV_ILLEGAL_INPUT;                                   \
 120             break;                                                            \
 121           }                                                                   \
 122         curcs = sb;                                                           \
 123         ++inptr;                                                              \
 124         continue;                                                             \
 125       }                                                                       \
 126                                                                               \
 127     if (curcs == sb)                                                          \
 128       {                                                                       \
 129         /* Use the UCS4 table for single byte.  */                            \
 130         ch = __ibm937sb_to_ucs4[ch];                                          \
 131         if (__builtin_expect (ch, L'\1') == L'\0' && *inptr != '\0')          \
 132           {                                                                   \
 133             /* This is an illegal character.  */                              \
 134             if (! ignore_errors_p ())                                         \
 135               {                                                               \
 136                 result = __GCONV_ILLEGAL_INPUT;                               \
 137                 break;                                                        \
 138               }                                                               \
 139             ++*irreversible;                                                  \
 140           }                                                                   \
 141         else                                                                  \
 142           {                                                                   \
 143             put32 (outptr, ch);                                               \
 144             outptr += 4;                                                      \
 145           }                                                                   \
 146         ++inptr;                                                              \
 147       }                                                                       \
 148     else                                                                      \
 149       {                                                                       \
 150         /* Use the IBM937 table for double byte.  */                          \
 151                                                                               \
 152         assert (curcs == db);                                                 \
 153                                                                               \
 154         ch = ibm937db_to_ucs4(inptr[0], inptr[1]);                            \
 155         if (__builtin_expect (ch, L'\1') == L'\0' && *inptr != '\0')          \
 156           {                                                                   \
 157             /* This is an illegal character.  */                              \
 158             if (! ignore_errors_p ())                                         \
 159               {                                                               \
 160                 result = __GCONV_ILLEGAL_INPUT;                               \
 161                 break;                                                        \
 162               }                                                               \
 163             ++*irreversible;                                                  \
 164           }                                                                   \
 165         else                                                                  \
 166           {                                                                   \
 167             put32 (outptr, ch);                                               \
 168             outptr += 4;                                                      \
 169           }                                                                   \
 170         inptr += 2;                                                           \
 171       }                                                                       \
 172   }
 173 #define LOOP_NEED_FLAGS
 174 #define EXTRA_LOOP_DECLS        , int *curcsp
 175 #define INIT_PARAMS             int curcs = *curcsp & ~7
 176 #define UPDATE_PARAMS           *curcsp = curcs
 177 #include <iconv/loop.c>
 178
 179 /* Next, define the other direction.  */
 180 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 181 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 182 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 183 #define LOOPFCT                 TO_LOOP
 184 #define BODY \
 185   {                                                                           \
 186     uint32_t ch = get32 (inptr);                                              \
 187     const char *cp;                                                           \
 188                                                                               \
 189     /* Use the UCS4 table for single byte.  */                                \
 190     cp = __ucs4_to_ibm937sb[ch];                                              \
 191     if (__builtin_expect (ch >= (sizeof (__ucs4_to_ibm937sb)                  \
 192                                  / sizeof (__ucs4_to_ibm937sb[0])), 0)        \
 193         || (__builtin_expect (cp[0], '\1') == '\0' && ch != 0))               \
 194       {                                                                       \
 195         /* Use the UCS4 table for double byte.  */                            \
 196         cp = __ucs4_to_ibm937db[ch];                                          \
 197         if (__builtin_expect (ch >= (sizeof (__ucs4_to_ibm937db)              \
 198                                      / sizeof (__ucs4_to_ibm937db[0])), 0)    \
 199             || __builtin_expect (cp[0], '\1') == '\0')                        \
 200           {                                                                   \
 201             /* This is an illegal character.  */                              \
 202             if (! ignore_errors_p ())                                         \
 203               {                                                               \
 204                 result = __GCONV_ILLEGAL_INPUT;                               \
 205                 break;                                                        \
 206               }                                                               \
 207             ++*irreversible;                                                  \
 208           }                                                                   \
 209         else                                                                  \
 210           {                                                                   \
 211             if (curcs == sb)                                                  \
 212               {                                                               \
 213                 *outptr++ = SO;                                               \
 214                 curcs = db;                                                   \
 215               }                                                               \
 216             if (__builtin_expect (outptr + 1 >= outend, 0))                   \
 217               {                                                               \
 218                 result = __GCONV_FULL_OUTPUT;                                 \
 219                 break;                                                        \
 220               }                                                               \
 221             *outptr++ = cp[0];                                                \
 222             *outptr++ = cp[1];                                                \
 223           }                                                                   \
 224       }                                                                       \
 225     else                                                                      \
 226       {                                                                       \
 227         if (curcs == db)                                                      \
 228           {                                                                   \
 229             *outptr++ = SI;                                                   \
 230             curcs = sb;                                                       \
 231             if (__builtin_expect (outptr == outend, 0))                       \
 232               {                                                               \
 233                 result = __GCONV_FULL_OUTPUT;                                 \
 234                 break;                                                        \
 235               }                                                               \
 236           }                                                                   \
 237         *outptr++ = cp[0];                                                    \
 238       }                                                                       \
 239                                                                               \
 240     /* Now that we wrote the output increment the input pointer.  */          \
 241     inptr += 4;                                                               \
 242   }
 243 #define LOOP_NEED_FLAGS
 244 #define EXTRA_LOOP_DECLS        , int *curcsp
 245 #define INIT_PARAMS             int curcs = *curcsp & ~7
 246 #define UPDATE_PARAMS           *curcsp = curcs
 247 #include <iconv/loop.c>
 248
 249 /* Now define the toplevel functions.  */
 250 #include <iconv/skeleton.c>