/* Conversion module for UTF-16.
- Copyright (C) 1999, 2000 Free Software Foundation, Inc.
+ Copyright (C) 1999-2014 Free Software Foundation, Inc.
This file is part of the GNU C Library.
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1999.
The GNU C Library is free software; you can redistribute it and/or
- modify it under the terms of the GNU Library General Public License as
- published by the Free Software Foundation; either version 2 of the
- License, or (at your option) any later version.
+ modify it under the terms of the GNU Lesser General Public
+ License as published by the Free Software Foundation; either
+ version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- Library General Public License for more details.
+ Lesser General Public License for more details.
- You should have received a copy of the GNU Library General Public
- License along with the GNU C Library; see the file COPYING.LIB. If not,
- write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- Boston, MA 02111-1307, USA. */
+ You should have received a copy of the GNU Lesser General Public
+ License along with the GNU C Library; if not, see
+ <http://www.gnu.org/licenses/>. */
#include <byteswap.h>
+#include <dlfcn.h>
#include <gconv.h>
#include <stddef.h>
#include <stdint.h>
#define MIN_NEEDED_FROM 2
#define MAX_NEEDED_FROM 4
#define MIN_NEEDED_TO 4
+#define ONE_DIRECTION 0
#define FROM_DIRECTION (dir == from_utf16)
#define PREPARE_LOOP \
enum direction dir = ((struct utf16_data *) step->__data)->dir; \
enum variant var = ((struct utf16_data *) step->__data)->var; \
- int swap = ((struct utf16_data *) step->__data)->swap; \
- if (FROM_DIRECTION || var == UTF_16) \
+ if (__glibc_unlikely (data->__invocation_counter == 0)) \
{ \
- if (data->__invocation_counter == 0) \
+ if (var == UTF_16) \
{ \
- /* We have to find out which byte order the file is encoded in. */ \
- if (inptr + 2 > inend) \
- return __GCONV_EMPTY_INPUT; \
+ if (FROM_DIRECTION) \
+ { \
+ /* We have to find out which byte order the file is \
+ encoded in. */ \
+ if (inptr + 2 > inend) \
+ return (inptr == inend \
+ ? __GCONV_EMPTY_INPUT : __GCONV_INCOMPLETE_INPUT); \
\
- if (get16u (inptr) == BOM) \
- /* Simply ignore the BOM character. */ \
- inptr += 2; \
- else if (get16u (inptr) == BOM_OE) \
+ if (get16u (inptr) == BOM) \
+ /* Simply ignore the BOM character. */ \
+ *inptrp = inptr += 2; \
+ else if (get16u (inptr) == BOM_OE) \
+ { \
+ data->__flags |= __GCONV_SWAP; \
+ *inptrp = inptr += 2; \
+ } \
+ } \
+ else if (!FROM_DIRECTION && !data->__internal_use) \
{ \
- ((struct utf16_data *) step->__data)->swap = 1; \
- inptr += 2; \
+ /* Emit the Byte Order Mark. */ \
+ if (__glibc_unlikely (outbuf + 2 > outend)) \
+ return __GCONV_FULL_OUTPUT; \
+ \
+ put16u (outbuf, BOM); \
+ outbuf += 2; \
} \
} \
+ else if ((var == UTF_16LE && BYTE_ORDER == BIG_ENDIAN) \
+ || (var == UTF_16BE && BYTE_ORDER == LITTLE_ENDIAN)) \
+ data->__flags |= __GCONV_SWAP; \
} \
- else if (!FROM_DIRECTION && var == UTF_16 && !data->__internal_use \
- && data->__invocation_counter == 0) \
- { \
- /* Emit the Byte Order Mark. */ \
- if (outbuf + 2 > outend) \
- return __GCONV_FULL_OUTPUT; \
- \
- put16u (outbuf, BOM); \
- outbuf += 2; \
- }
-#define EXTRA_LOOP_ARGS , var, data, swap
+ const int swap = data->__flags & __GCONV_SWAP;
+#define EXTRA_LOOP_ARGS , swap
/* Direction of the transformation. */
{
enum direction dir;
enum variant var;
- int swap;
};
+extern int gconv_init (struct __gconv_step *step);
int
gconv_init (struct __gconv_step *step)
{
enum variant var = illegal_var;
int result;
- if (__strcasecmp (step->__from_name, "UTF-16") == 0)
+ if (__strcasecmp (step->__from_name, "UTF-16//") == 0)
{
dir = from_utf16;
var = UTF_16;
}
- else if (__strcasecmp (step->__to_name, "UTF-16") == 0)
+ else if (__strcasecmp (step->__to_name, "UTF-16//") == 0)
{
dir = to_utf16;
var = UTF_16;
}
- else if (__strcasecmp (step->__from_name, "UTF-16BE") == 0)
+ else if (__strcasecmp (step->__from_name, "UTF-16BE//") == 0)
{
dir = from_utf16;
var = UTF_16BE;
}
- else if (__strcasecmp (step->__to_name, "UTF-16BE") == 0)
+ else if (__strcasecmp (step->__to_name, "UTF-16BE//") == 0)
{
dir = to_utf16;
var = UTF_16BE;
}
- else if (__strcasecmp (step->__from_name, "UTF-16LE") == 0)
+ else if (__strcasecmp (step->__from_name, "UTF-16LE//") == 0)
{
dir = from_utf16;
var = UTF_16LE;
}
- else if (__strcasecmp (step->__to_name, "UTF-16LE") == 0)
+ else if (__strcasecmp (step->__to_name, "UTF-16LE//") == 0)
{
dir = to_utf16;
var = UTF_16LE;
}
result = __GCONV_NOCONV;
- if (dir != illegal_dir)
+ if (__builtin_expect (dir, to_utf16) != illegal_dir)
{
new_data = (struct utf16_data *) malloc (sizeof (struct utf16_data));
{
new_data->dir = dir;
new_data->var = var;
- new_data->swap = ((var == UTF_16LE && BYTE_ORDER == BIG_ENDIAN)
- || (var == UTF_16BE
- && BYTE_ORDER == LITTLE_ENDIAN));
step->__data = new_data;
if (dir == from_utf16)
{
step->__min_needed_from = MIN_NEEDED_FROM;
- step->__max_needed_from = MIN_NEEDED_FROM;
+ step->__max_needed_from = MAX_NEEDED_FROM;
step->__min_needed_to = MIN_NEEDED_TO;
step->__max_needed_to = MIN_NEEDED_TO;
}
step->__min_needed_from = MIN_NEEDED_TO;
step->__max_needed_from = MIN_NEEDED_TO;
step->__min_needed_to = MIN_NEEDED_FROM;
- step->__max_needed_to = MIN_NEEDED_FROM;
+ step->__max_needed_to = MAX_NEEDED_FROM;
}
step->__stateful = 0;
}
+extern void gconv_end (struct __gconv_step *data);
void
gconv_end (struct __gconv_step *data)
{
{ \
uint32_t c = get32 (inptr); \
\
+ if (__glibc_unlikely (c >= 0xd800 && c < 0xe000)) \
+ { \
+ /* Surrogate characters in UCS-4 input are not valid. \
+ We must catch this. If we let surrogates pass through, \
+ attackers could make a security hole exploit by \
+ synthesizing any desired plane 1-16 character. */ \
+ result = __GCONV_ILLEGAL_INPUT; \
+ if (! ignore_errors_p ()) \
+ break; \
+ inptr += 4; \
+ ++*irreversible; \
+ continue; \
+ } \
+ \
if (swap) \
{ \
- if (c >= 0x10000) \
+ if (__glibc_unlikely (c >= 0x10000)) \
{ \
- if (c >= 0x110000) \
+ if (__glibc_unlikely (c >= 0x110000)) \
{ \
- result = __GCONV_ILLEGAL_INPUT; \
- break; \
+ STANDARD_TO_LOOP_ERR_HANDLER (4); \
} \
\
/* Generate a surrogate character. */ \
- if (NEED_LENGTH_TEST && outptr + 4 > outend) \
+ if (__glibc_unlikely (outptr + 4 > outend)) \
{ \
/* Overflow in the output buffer. */ \
result = __GCONV_FULL_OUTPUT; \
} \
else \
{ \
- if (c >= 0x10000) \
+ if (__glibc_unlikely (c >= 0x10000)) \
{ \
- if (c >= 0x110000) \
+ if (__glibc_unlikely (c >= 0x110000)) \
{ \
- result = __GCONV_ILLEGAL_INPUT; \
- break; \
+ STANDARD_TO_LOOP_ERR_HANDLER (4); \
} \
\
/* Generate a surrogate character. */ \
- if (NEED_LENGTH_TEST && outptr + 4 > outend) \
+ if (__glibc_unlikely (outptr + 4 > outend)) \
{ \
/* Overflow in the output buffer. */ \
result = __GCONV_FULL_OUTPUT; \
outptr += 2; \
inptr += 4; \
}
+#define LOOP_NEED_FLAGS
#define EXTRA_LOOP_DECLS \
- , enum variant var, struct __gconv_step_data *step_data, int swap
+ , int swap
#include <iconv/loop.c>
{ \
u1 = bswap_16 (u1); \
\
- if (u1 < 0xd800 || u1 > 0xdfff) \
+ if (__builtin_expect (u1 < 0xd800, 1) || u1 > 0xdfff) \
{ \
/* No surrogate. */ \
- put32 (outptr, u1); \
+ put32 (outptr, u1); \
inptr += 2; \
} \
else \
\
/* It's a surrogate character. At least the first word says \
it is. */ \
- if (NEED_LENGTH_TEST && inptr + 4 > inend) \
+ if (__glibc_unlikely (inptr + 4 > inend)) \
{ \
/* We don't have enough input for another complete input \
character. */ \
\
inptr += 2; \
u2 = bswap_16 (get16 (inptr)); \
- if (u2 < 0xdc00 || u2 >= 0xdfff) \
+ if (__builtin_expect (u2 < 0xdc00, 0) \
+ || __builtin_expect (u2 > 0xdfff, 0)) \
{ \
/* This is no valid second word for a surrogate. */ \
- result = __GCONV_ILLEGAL_INPUT; \
inptr -= 2; \
- break; \
+ STANDARD_FROM_LOOP_ERR_HANDLER (2); \
} \
\
put32 (outptr, ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00)); \
} \
else \
{ \
- if (u1 < 0xd800 || u1 > 0xdfff) \
+ if (__builtin_expect (u1 < 0xd800, 1) || u1 > 0xdfff) \
{ \
/* No surrogate. */ \
put32 (outptr, u1); \
} \
else \
{ \
- uint16_t u2; \
- \
/* It's a surrogate character. At least the first word says \
it is. */ \
- if (NEED_LENGTH_TEST && inptr + 4 > inend) \
+ if (__glibc_unlikely (inptr + 4 > inend)) \
{ \
/* We don't have enough input for another complete input \
character. */ \
} \
\
inptr += 2; \
- u2 = get16 (inptr); \
- if (u2 < 0xdc00 || u2 >= 0xdfff) \
+ uint16_t u2 = get16 (inptr); \
+ if (__builtin_expect (u2 < 0xdc00, 0) \
+ || __builtin_expect (u2 > 0xdfff, 0)) \
{ \
/* This is no valid second word for a surrogate. */ \
- result = __GCONV_ILLEGAL_INPUT; \
inptr -= 2; \
- break; \
+ STANDARD_FROM_LOOP_ERR_HANDLER (2); \
} \
\
put32 (outptr, ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00)); \
} \
outptr += 4; \
}
+#define LOOP_NEED_FLAGS
#define EXTRA_LOOP_DECLS \
- , enum variant var, struct __gconv_step_data *step_data, int swap
+ , int swap
#include <iconv/loop.c>