iconvdata/ibm932.c

   1 /* Conversion from and to IBM932.
   2    Copyright (C) 2000-2024 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 <stdint.h>
  21 #include <stdbool.h>
  22 #include "ibm932.h"
  23
  24 #define FROM    0
  25 #define TO      1
  26
  27 /* Definitions used in the body of the `gconv' function.  */
  28 #define CHARSET_NAME    "IBM932//"
  29 #define FROM_LOOP       from_ibm932
  30 #define TO_LOOP         to_ibm932
  31 #define ONE_DIRECTION   0
  32
  33 /* Definitions of initialization and destructor function.  */
  34 #define DEFINE_INIT     1
  35 #define DEFINE_FINI     1
  36
  37 #define MIN_NEEDED_FROM 1
  38 #define MAX_NEEDED_FROM 2
  39 #define MIN_NEEDED_TO   4
  40
  41 /* First, define the conversion function from IBM-932 to UCS4.  */
  42 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
  43 #define MAX_NEEDED_INPUT        MAX_NEEDED_FROM
  44 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
  45 #define LOOPFCT                 FROM_LOOP
  46 #define BODY \
  47   {                                                                           \
  48     const struct gap *rp2 = __ibm932db_to_ucs4_idx;                           \
  49     uint32_t ch = *inptr;                                                     \
  50     uint32_t res;                                                             \
  51                                                                               \
  52     if (__builtin_expect (ch == 0x80, 0)                                      \
  53         || __builtin_expect (ch == 0xa0, 0)                                   \
  54         || __builtin_expect (ch == 0xfd, 0)                                   \
  55         || __builtin_expect (ch == 0xfe, 0)                                   \
  56         || __builtin_expect (ch == 0xff, 0))                                  \
  57       {                                                                       \
  58         /* This is an illegal character.  */                                  \
  59         STANDARD_FROM_LOOP_ERR_HANDLER (1);                                   \
  60       }                                                                       \
  61                                                                               \
  62     /* Use the IBM932 table for single byte.  */                              \
  63     res = __ibm932sb_to_ucs4[ch];                                             \
  64     if (__builtin_expect (res == 0, 0) && ch != 0)                            \
  65       {                                                                       \
  66         /* Use the IBM932 table for double byte.  */                          \
  67         if (__glibc_unlikely (inptr + 1 >= inend))                            \
  68           {                                                                   \
  69             /* The second character is not available.                         \
  70                Store the intermediate result.  */                             \
  71             result = __GCONV_INCOMPLETE_INPUT;                                \
  72             break;                                                            \
  73           }                                                                   \
  74                                                                               \
  75         ch = (ch * 0x100) + inptr[1];                                         \
  76         /* ch was less than 0xfd.  */                                         \
  77         assert (ch < 0xfd00);                                                 \
  78         while (ch > rp2->end)                                                 \
  79           ++rp2;                                                              \
  80                                                                               \
  81         if (__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 == 0xa5)                                                      \
  98           res = 0x5c;                                                         \
  99         else if (res == 0x203e)                                               \
 100           res = 0x7e;                                                         \
 101         put32 (outptr, res);                                                  \
 102         outptr += 4;                                                          \
 103         inptr++;                                                              \
 104       }                                                                       \
 105   }
 106 #define LOOP_NEED_FLAGS
 107 #define ONEBYTE_BODY \
 108   {                                                                           \
 109     if (c == 0x80 || c == 0xa0 || c >= 0xfd)                                  \
 110       return WEOF;                                                            \
 111     uint32_t res = __ibm932sb_to_ucs4[c];                                     \
 112     if (res == 0 && c != 0)                                                   \
 113       return WEOF;                                                            \
 114     if (res == 0xa5)                                                          \
 115       res = 0x5c;                                                             \
 116     else if (res == 0x203e)                                                   \
 117       res = 0x7e;                                                             \
 118     return res;                                                               \
 119   }
 120 #include <iconv/loop.c>
 121
 122 /* Next, define the other direction.  */
 123 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 124 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 125 #define MAX_NEEDED_OUTPUT       MAX_NEEDED_FROM
 126 #define LOOPFCT                 TO_LOOP
 127 #define BODY \
 128   {                                                                           \
 129     const struct gap *rp = __ucs4_to_ibm932sb_idx;                            \
 130     unsigned char sc;                                                         \
 131     uint32_t ch = get32 (inptr);                                              \
 132     bool found = true;                                                        \
 133     uint32_t i;                                                               \
 134     uint32_t low;                                                             \
 135     uint32_t high;                                                            \
 136     uint16_t pccode;                                                          \
 137                                                                               \
 138     if (__glibc_unlikely (ch >= 0xffff))                                      \
 139       {                                                                       \
 140         UNICODE_TAG_HANDLER (ch, 4);                                          \
 141         rp = NULL;                                                            \
 142       }                                                                       \
 143     else                                                                      \
 144       while (ch > rp->end)                                                    \
 145         ++rp;                                                                 \
 146                                                                               \
 147     /* Use the UCS4 table for single byte.  */                                \
 148     if (__builtin_expect (rp == NULL, 0)                                      \
 149         || __builtin_expect (ch < rp->start, 0)                               \
 150         || (sc = __ucs4_to_ibm932sb[ch + rp->idx],                            \
 151         __builtin_expect (sc, '\1') == '\0' && ch != L'\0'))                  \
 152       {                                                                       \
 153                                                                               \
 154         /* Use the UCS4 table for double byte.  */                            \
 155         found = false;                                                        \
 156         low = 0;                                                              \
 157         high = (sizeof (__ucs4_to_ibm932db) >> 1)                             \
 158                 / sizeof (__ucs4_to_ibm932db[0][FROM]);                       \
 159         pccode = ch;                                                          \
 160         if (__glibc_likely (rp != NULL))                                      \
 161           while (low < high)                                                  \
 162             {                                                                 \
 163               i = (low + high) >> 1;                                          \
 164               if (pccode < __ucs4_to_ibm932db[i][FROM])                       \
 165                 high = i;                                                     \
 166               else if (pccode > __ucs4_to_ibm932db[i][FROM])                  \
 167                 low = i + 1;                                                  \
 168               else                                                            \
 169                 {                                                             \
 170                   pccode = __ucs4_to_ibm932db[i][TO];                         \
 171                   found = true;                                               \
 172                   break;                                                      \
 173                 }                                                             \
 174             }                                                                 \
 175         if (found)                                                            \
 176           {                                                                   \
 177             if (__glibc_unlikely (outptr + 2 > outend))                       \
 178               {                                                               \
 179                 result = __GCONV_FULL_OUTPUT;                                 \
 180                 break;                                                        \
 181               }                                                               \
 182             *outptr++ = pccode >> 8 & 0xff;                                   \
 183             *outptr++ = pccode & 0xff;                                        \
 184           }                                                                   \
 185         else                                                                  \
 186           {                                                                   \
 187             /* This is an illegal character.  */                              \
 188             STANDARD_TO_LOOP_ERR_HANDLER (4);                                 \
 189           }                                                                   \
 190       }                                                                       \
 191     else                                                                      \
 192       {                                                                       \
 193         if (__glibc_unlikely (outptr + 1 > outend))                           \
 194           {                                                                   \
 195             result = __GCONV_FULL_OUTPUT;                                     \
 196             break;                                                            \
 197           }                                                                   \
 198         if (ch == 0x5c)                                                       \
 199           *outptr++ = 0x5c;                                                   \
 200         else if (ch == 0x7e)                                                  \
 201           *outptr++ = 0x7e;                                                   \
 202         else                                                                  \
 203           *outptr++ = sc;                                                     \
 204       }                                                                       \
 205                                                                               \
 206     /* Now that we wrote the output increment the input pointer.  */          \
 207     inptr += 4;                                                               \
 208   }
 209 #define LOOP_NEED_FLAGS
 210 #include <iconv/loop.c>
 211
 212 /* Now define the toplevel functions.  */
 213 #include <iconv/skeleton.c>