iconvdata/utf-32.c

   1 /* Conversion module for UTF-32.
   2    Copyright (C) 1999, 2000-2002 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, write to the Free
  17    Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
  18    02111-1307 USA.  */
  19
  20 #include <byteswap.h>
  21 #include <dlfcn.h>
  22 #include <gconv.h>
  23 #include <stddef.h>
  24 #include <stdint.h>
  25 #include <stdlib.h>
  26 #include <string.h>
  27
  28 /* This is the Byte Order Mark character (BOM).  */
  29 #define BOM     0x0000feffu
  30 /* And in the other byte order.  */
  31 #define BOM_OE  0xfffe0000u
  32
  33
  34 /* Definitions used in the body of the `gconv' function.  */
  35 #define FROM_LOOP               from_utf32_loop
  36 #define TO_LOOP                 to_utf32_loop
  37 #define DEFINE_INIT             0
  38 #define DEFINE_FINI             0
  39 #define MIN_NEEDED_FROM         4
  40 #define MIN_NEEDED_TO           4
  41 #define FROM_DIRECTION          (dir == from_utf32)
  42 #define PREPARE_LOOP \
  43   enum direction dir = ((struct utf32_data *) step->__data)->dir;             \
  44   enum variant var = ((struct utf32_data *) step->__data)->var;               \
  45   int swap;                                                                   \
  46   if (FROM_DIRECTION && var == UTF_32)                                        \
  47     {                                                                         \
  48       if (data->__invocation_counter == 0)                                    \
  49         {                                                                     \
  50           /* We have to find out which byte order the file is encoded in.  */ \
  51           if (inptr + 4 > inend)                                              \
  52             return __GCONV_EMPTY_INPUT;                                       \
  53                                                                               \
  54           if (get32u (inptr) == BOM)                                          \
  55             /* Simply ignore the BOM character.  */                           \
  56             *inptrp = inptr += 4;                                             \
  57           else if (get32u (inptr) == BOM_OE)                                  \
  58             {                                                                 \
  59               ((struct utf32_data *) step->__data)->swap = 1;                 \
  60               *inptrp = inptr += 4;                                           \
  61             }                                                                 \
  62         }                                                                     \
  63     }                                                                         \
  64   else if (!FROM_DIRECTION && var == UTF_32 && !data->__internal_use          \
  65            && data->__invocation_counter == 0)                                \
  66     {                                                                         \
  67       /* Emit the Byte Order Mark.  */                                        \
  68       if (__builtin_expect (outbuf + 4 > outend, 0))                          \
  69         return __GCONV_FULL_OUTPUT;                                           \
  70                                                                               \
  71       put32u (outbuf, BOM);                                                   \
  72       outbuf += 4;                                                            \
  73     }                                                                         \
  74   swap = ((struct utf32_data *) step->__data)->swap;
  75 #define EXTRA_LOOP_ARGS         , var, swap
  76
  77
  78 /* Direction of the transformation.  */
  79 enum direction
  80 {
  81   illegal_dir,
  82   to_utf32,
  83   from_utf32
  84 };
  85
  86 enum variant
  87 {
  88   illegal_var,
  89   UTF_32,
  90   UTF_32LE,
  91   UTF_32BE
  92 };
  93
  94 struct utf32_data
  95 {
  96   enum direction dir;
  97   enum variant var;
  98   int swap;
  99 };
 100
 101
 102 extern int gconv_init (struct __gconv_step *step);
 103 int
 104 gconv_init (struct __gconv_step *step)
 105 {
 106   /* Determine which direction.  */
 107   struct utf32_data *new_data;
 108   enum direction dir = illegal_dir;
 109   enum variant var = illegal_var;
 110   int result;
 111
 112   if (__strcasecmp (step->__from_name, "UTF-32//") == 0)
 113     {
 114       dir = from_utf32;
 115       var = UTF_32;
 116     }
 117   else if (__strcasecmp (step->__to_name, "UTF-32//") == 0)
 118     {
 119       dir = to_utf32;
 120       var = UTF_32;
 121     }
 122   else if (__strcasecmp (step->__from_name, "UTF-32BE//") == 0)
 123     {
 124       dir = from_utf32;
 125       var = UTF_32BE;
 126     }
 127   else if (__strcasecmp (step->__to_name, "UTF-32BE//") == 0)
 128     {
 129       dir = to_utf32;
 130       var = UTF_32BE;
 131     }
 132   else if (__strcasecmp (step->__from_name, "UTF-32LE//") == 0)
 133     {
 134       dir = from_utf32;
 135       var = UTF_32LE;
 136     }
 137   else if (__strcasecmp (step->__to_name, "UTF-32LE//") == 0)
 138     {
 139       dir = to_utf32;
 140       var = UTF_32LE;
 141     }
 142
 143   result = __GCONV_NOCONV;
 144   if (__builtin_expect (dir, to_utf32) != illegal_dir)
 145     {
 146       new_data = (struct utf32_data *) malloc (sizeof (struct utf32_data));
 147
 148       result = __GCONV_NOMEM;
 149       if (new_data != NULL)
 150         {
 151           new_data->dir = dir;
 152           new_data->var = var;
 153           new_data->swap = ((var == UTF_32LE && BYTE_ORDER == BIG_ENDIAN)
 154                             || (var == UTF_32BE
 155                                 && BYTE_ORDER == LITTLE_ENDIAN));
 156           step->__data = new_data;
 157
 158           if (dir == from_utf32)
 159             {
 160               step->__min_needed_from = MIN_NEEDED_FROM;
 161               step->__max_needed_from = MIN_NEEDED_FROM;
 162               step->__min_needed_to = MIN_NEEDED_TO;
 163               step->__max_needed_to = MIN_NEEDED_TO;
 164             }
 165           else
 166             {
 167               step->__min_needed_from = MIN_NEEDED_TO;
 168               step->__max_needed_from = MIN_NEEDED_TO;
 169               step->__min_needed_to = MIN_NEEDED_FROM;
 170               step->__max_needed_to = MIN_NEEDED_FROM;
 171             }
 172
 173           step->__stateful = 0;
 174
 175           result = __GCONV_OK;
 176         }
 177     }
 178
 179   return result;
 180 }
 181
 182
 183 extern void gconv_end (struct __gconv_step *data);
 184 void
 185 gconv_end (struct __gconv_step *data)
 186 {
 187   free (data->__data);
 188 }
 189
 190
 191 /* Convert from the internal (UCS4-like) format to UTF-32.  */
 192 #define MIN_NEEDED_INPUT        MIN_NEEDED_TO
 193 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_FROM
 194 #define LOOPFCT                 TO_LOOP
 195 #define BODY \
 196   {                                                                           \
 197     uint32_t c = get32 (inptr);                                               \
 198                                                                               \
 199     if (__builtin_expect (c >= 0x110000, 0))                                  \
 200       {                                                                       \
 201         STANDARD_TO_LOOP_ERR_HANDLER (4);                                     \
 202       }                                                                       \
 203     else if (__builtin_expect (c >= 0xd800 && c < 0xe000, 0))                 \
 204       {                                                                       \
 205         /* Surrogate characters in UCS-4 input are not valid.                 \
 206            We must catch this.  If we let surrogates pass through,            \
 207            attackers could make a security hole exploit by                    \
 208            generating "irregular UTF-32" sequences.  */                       \
 209         result = __GCONV_ILLEGAL_INPUT;                                       \
 210         if (! ignore_errors_p ())                                             \
 211           break;                                                              \
 212         inptr += 4;                                                           \
 213         ++*irreversible;                                                      \
 214         continue;                                                             \
 215       }                                                                       \
 216                                                                               \
 217     if (swap)                                                                 \
 218       put32 (outptr, bswap_32 (c));                                           \
 219     else                                                                      \
 220       put32 (outptr, c);                                                      \
 221                                                                               \
 222     outptr += 4;                                                              \
 223     inptr += 4;                                                               \
 224   }
 225 #define LOOP_NEED_FLAGS
 226 #define EXTRA_LOOP_DECLS \
 227         , enum variant var, int swap
 228 #include <iconv/loop.c>
 229
 230
 231 /* Convert from UTF-32 to the internal (UCS4-like) format.  */
 232 #define MIN_NEEDED_INPUT        MIN_NEEDED_FROM
 233 #define MIN_NEEDED_OUTPUT       MIN_NEEDED_TO
 234 #define LOOPFCT                 FROM_LOOP
 235 #define BODY \
 236   {                                                                           \
 237     uint32_t u1 = get32 (inptr);                                              \
 238                                                                               \
 239     if (swap)                                                                 \
 240       u1 = bswap_32 (u1);                                                     \
 241                                                                               \
 242     if (__builtin_expect (u1 >= 0x110000, 0))                                 \
 243       {                                                                       \
 244         /* This is illegal.  */                                               \
 245         STANDARD_FROM_LOOP_ERR_HANDLER (4);                                   \
 246       }                                                                       \
 247                                                                               \
 248     put32 (outptr, u1);                                                       \
 249     inptr += 4;                                                               \
 250     outptr += 4;                                                              \
 251   }
 252 #define LOOP_NEED_FLAGS
 253 #define EXTRA_LOOP_DECLS \
 254         , enum variant var, int swap
 255 #include <iconv/loop.c>
 256
 257
 258 /* Now define the toplevel functions.  */
 259 #include <iconv/skeleton.c>