iconvdata/ibm932.c

   1 /* Conversion from and to IBM932.
   2    Copyright (C) 2000-2002, 2004 Free Software Foundation, Inc.
   3    This file is part of the GNU C Library.
   4    Contributed by Masahide Washizawa <washi@jp.ibm.com>, 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 Lesser General Public
   8    License as published by the Free Software Foundation; either
   9    version 2.1 of the 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    Lesser General Public License for more details.
  15
  16    You should have received a copy of the GNU Lesser General Public
  17    License along with the GNU C Library; if not, write to the Free
  18    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  19    02111-1307 USA.  */
  20
  21 #include <dlfcn.h>
  22 #include <stdint.h>
  23 #include <stdbool.h>
  24 #include "ibm932.h"
  25
  26 #define FROM    0
  27 #define TO      1
  28
  29 /* Definitions used in the body of the `gconv' function.  */
  30 #define CHARSET_NAME    "IBM932//"
  31 #define FROM_LOOP       from_ibm932
  32 #define TO_LOOP         to_ibm932
  33
  34 /* Definitions of initialization and destructor function.  */
  35 #define DEFINE_INIT     1
  36 #define DEFINE_FINI     1
  37
  38 #define MIN_NEEDED_FROM 1
  39 #define MAX_NEEDED_FROM 2
  40 #define MIN_NEEDED_TO   4
  41
  42 /* First, define the conversion function from IBM-932 to UCS4.  */
  43 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  44 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  45 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  46 #define LOOPFCT                 FROM_LOOP
  47 #define BODY \
  48   {                                                                           \
  49     const struct gap *rp2 = __ibm932db_to_ucs4_idx;                           \
  50     uint32_t ch = *inptr;                                                     \
  51     uint32_t res;                                                             \
  52                                                                               \
  53     if (__builtin_expect (ch == 0x80, 0)                                      \
  54         || __builtin_expect (ch == 0xa0, 0)                                   \
  55         || __builtin_expect (ch == 0xfd, 0)                                   \
  56         || __builtin_expect (ch == 0xfe, 0)                                   \
  57         || __builtin_expect (ch == 0xff, 0))                                  \
  58       {                                                                       \
  59         /* This is an illegal character.  */                                  \
  60         STANDARD_FROM_LOOP_ERR_HANDLER (1);                                   \
  61       }                                                                       \
  62                                                                               \
  63     /* Use the IBM932 table for single byte.  */                              \
  64     res = __ibm932sb_to_ucs4[ch];                                             \
  65     if (__builtin_expect (res == 0, 0) && ch != 0)                            \
  66       {                                                                       \
  67         /* Use the IBM932 table for double byte.  */                          \
  68         if (__builtin_expect (inptr + 1 >= inend, 0))                         \
  69           {                                                                   \
  70             /* The second character is not available.                         \
  71                Store the intermediate result.  */                             \
  72             result = __GCONV_INCOMPLETE_INPUT;                                \
  73             break;                                                            \
  74           }                                                                   \
  75                                                                               \
  76         ch = (ch * 0x100) + inptr[1];                                         \
  77         while (ch > rp2->end)                                                 \
  78           ++rp2;                                                              \
  79                                                                               \
  80         if (__builtin_expect (rp2 == NULL, 0)                                 \
  81             || __builtin_expect (ch < rp2->start, 0)                          \
  82             || (res = __ibm932db_to_ucs4[ch + rp2->idx],                      \
  83             __builtin_expect (res, '\1') == 0 && ch !=0))                     \
  84           {                                                                   \
  85             /* This is an illegal character.  */                              \
  86             STANDARD_FROM_LOOP_ERR_HANDLER (2);                               \
  87           }                                                                   \
  88         else                                                                  \
  89           {                                                                   \
  90             put32 (outptr, res);                                              \
  91             outptr += 4;                                                      \
  92             inptr += 2;                                                       \
  93           }                                                                   \
  94       }                                                                       \
  95     else                                                                      \
  96       {                                                                       \
  97         if (res == 0x1c)                                                      \
  98           res = 0x1a;                                                         \
  99         else if (res == 0x7f)                                                 \
 100           res = 0x1c;                                                         \
 101         else if (res == 0xa5)                                                 \
 102           res = 0x5c;                                                         \
 103         else if (res == 0x203e)                                               \
 104           res = 0x7e;                                                         \
 105         else if (res == 0x1a)                                                 \
 106           res = 0x7f;                                                         \
 107         put32 (outptr, res);                                                  \
 108         outptr += 4;                                                          \
 109         inptr++;                                                              \
 110       }                                                                       \
 111   }
 112 #define LOOP_NEED_FLAGS
 113 #define ONEBYTE_BODY \
 114   {                                                                           \
 115     if (c == 0x80 || c == 0xa0 || c >= 0xfd)                                  \
 116       return WEOF;                                                            \
 117     uint32_t res = __ibm932sb_to_ucs4[c];                                     \
 118     if (res == 0 && c != 0)                                                   \
 119       return WEOF;                                                            \
 120     if (res == 0x1c)                                                          \
 121       res = 0x1a;                                                             \
 122     else if (res == 0x7f)                                                     \
 123       res = 0x1c;                                                             \
 124     else if (res == 0xa5)                                                     \
 125       res = 0x5c;                                                             \
 126     else if (res == 0x203e)                                                   \
 127       res = 0x7e;                                                             \
 128     else if (res == 0x1a)                                                     \
 129       res = 0x7f;                                                             \
 130     return res;                                                               \
 131   }
 132 #include <iconv/loop.c>
 133
 134 /* Next, define the other direction.  */
 135 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 136 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 137 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 138 #define LOOPFCT                 TO_LOOP
 139 #define BODY \
 140   {                                                                           \
 141     const struct gap *rp = __ucs4_to_ibm932sb_idx;                            \
 142     unsigned char sc;                                                         \
 143     uint32_t ch = get32 (inptr);                                              \
 144     bool found = true;                                                        \
 145     uint32_t i;                                                               \
 146     uint32_t low;                                                             \
 147     uint32_t high;                                                            \
 148     uint16_t pccode;                                                          \
 149                                                                               \
 150     if (__builtin_expect (ch >= 0xffff, 0))                                   \
 151       {                                                                       \
 152         UNICODE_TAG_HANDLER (ch, 4);                                          \
 153         rp = NULL;                                                            \
 154       }                                                                       \
 155     else                                                                      \
 156       while (ch > rp->end)                                                    \
 157         ++rp;                                                                 \
 158                                                                               \
 159     /* Use the UCS4 table for single byte.  */                                \
 160     if (__builtin_expect (rp == NULL, 0)                                      \
 161         || __builtin_expect (ch < rp->start, 0)                               \
 162         || (sc = __ucs4_to_ibm932sb[ch + rp->idx],                            \
 163         __builtin_expect (sc, '\1') == '\0' && ch != L'\0'))                  \
 164       {                                                                       \
 165                                                                               \
 166         /* Use the UCS4 table for double byte.  */                            \
 167         found = false;                                                        \
 168         low = 0;                                                              \
 169         high = (sizeof (__ucs4_to_ibm932db) >> 1)                             \
 170                 / sizeof (__ucs4_to_ibm932db[0][FROM]);                       \
 171         pccode = ch;                                                          \
 172         if (__builtin_expect (rp != NULL, 1))                                 \
 173           while (low < high)                                                  \
 174             {                                                                 \
 175               i = (low + high) >> 1;                                          \
 176               if (pccode < __ucs4_to_ibm932db[i][FROM])                       \
 177                 high = i;                                                     \
 178               else if (pccode > __ucs4_to_ibm932db[i][FROM])                  \
 179                 low = i + 1;                                                  \
 180               else                                                            \
 181                 {                                                             \
 182                   pccode = __ucs4_to_ibm932db[i][TO];                         \
 183                   found = true;                                               \
 184                   break;                                                      \
 185                 }                                                             \
 186             }                                                                 \
 187         if (found)                                                            \
 188           {                                                                   \
 189             if (__builtin_expect (outptr + 2 > outend, 0))                    \
 190               {                                                               \
 191                 result = __GCONV_FULL_OUTPUT;                                 \
 192                 break;                                                        \
 193               }                                                               \
 194             *outptr++ = pccode >> 8 & 0xff;                                   \
 195             *outptr++ = pccode & 0xff;                                        \
 196           }                                                                   \
 197         else                                                                  \
 198           {                                                                   \
 199             /* This is an illegal character.  */                              \
 200             STANDARD_TO_LOOP_ERR_HANDLER (4);                                 \
 201           }                                                                   \
 202       }                                                                       \
 203     else                                                                      \
 204       {                                                                       \
 205         if (__builtin_expect (outptr + 1 > outend, 0))                        \
 206           {                                                                   \
 207             result = __GCONV_FULL_OUTPUT;                                     \
 208             break;                                                            \
 209           }                                                                   \
 210         if (ch == 0x5c)                                                       \
 211           *outptr++ = 0x5c;                                                   \
 212         else if (ch == 0x7e)                                                  \
 213           *outptr++ = 0x7e;                                                   \
 214         else                                                                  \
 215           *outptr++ = sc;                                                     \
 216       }                                                                       \
 217                                                                               \
 218     /* Now that we wrote the output increment the input pointer.  */          \
 219     inptr += 4;                                                               \
 220   }
 221 #define LOOP_NEED_FLAGS
 222 #include <iconv/loop.c>
 223
 224 /* Now define the toplevel functions.  */
 225 #include <iconv/skeleton.c>