mirror/glibc
2
0
Fork 0
mirror of git://sourceware.org/git/glibc.git synced 2024-11-21 01:12:26 +08:00
Code Issues Packages Projects Releases Wiki Activity
c7dcf594f4
glibc/iconvdata/utf-16.c
Florian Weimer 6cbf845fcd iconv: Preserve iconv -c error exit on invalid inputs (bug 32046) 
In several converters, a __GCONV_ILLEGAL_INPUT result gets overwritten
with __GCONV_FULL_OUTPUT.  As a result, iconv (the function) returns
E2BIG instead of EILSEQ.  The iconv program does not see the original
EILSEQ failure, does not recognize the invalid input, and may
incorrectly exit successfully.

To address this, a new __flags bit is used to indicate a sticky input
error state.  All __GCONV_ILLEGAL_INPUT results are replaced with a
function call that sets this new __GCONV_ENCOUNTERED_ILLEGAL_INPUT and
returns __GCONV_ILLEGAL_INPUT.  The iconv program checks for
__GCONV_ENCOUNTERED_ILLEGAL_INPUT and overrides the exit status.

The converter changes introducing __gconv_mark_illegal_input are
mostly mechanical, except for the res variable initialization in
iconvdata/iso-2022-jp.c: this error gets overwritten with __GCONV_OK
and other results in the following code.  If res ==
__GCONV_ILLEGAL_INPUT afterwards, STANDARD_TO_LOOP_ERR_HANDLER below
will handle it.

The __gconv_mark_illegal_input changes do not alter the errno value
set by the iconv function.  This is simpler to implement than
reviewing each __GCONV_FULL_OUTPUT result and adjust it not to
override a previous __GCONV_ILLEGAL_INPUT result.  Doing it that way
would also change some E2BIG errors in to EILSEQ errors, so it had to
be done conditionally (under a flag set by the iconv program only), to
avoid confusing buffer management in other applications.

Reviewed-by: DJ Delorie <dj@redhat.com>
2024-09-20 13:51:09 +02:00

377 lines
11 KiB
C
Raw Blame History

/* Conversion module for UTF-16.
Copyright (C) 1999-2024 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
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
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
<https://www.gnu.org/licenses/>. */
#include <byteswap.h>
#include <dlfcn.h>
#include <gconv.h>
#include <stddef.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
/* This is the Byte Order Mark character (BOM). */
#define BOM 0xfeff
/* And in the other byte order. */
#define BOM_OE 0xfffe
/* Definitions used in the body of the `gconv' function. */
#define FROM_LOOP from_utf16_loop
#define TO_LOOP to_utf16_loop
#define DEFINE_INIT 0
#define DEFINE_FINI 0
#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; \
if (__glibc_unlikely (data->__invocation_counter == 0)) \
{ \
if (var == UTF_16) \
{ \
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 (get16 (inptr) == BOM) \
/* Simply ignore the BOM character. */ \
*inptrp = inptr += 2; \
else if (get16 (inptr) == BOM_OE) \
{ \
data->__flags |= __GCONV_SWAP; \
*inptrp = inptr += 2; \
} \
} \
else if (!FROM_DIRECTION && !data->__internal_use) \
{ \
/* Emit the Byte Order Mark. */ \
if (__glibc_unlikely (outbuf + 2 > outend)) \
return __GCONV_FULL_OUTPUT; \
\
put16 (outbuf, BOM); \
outbuf += 2; \
} \
} \
else if ((var == UTF_16LE && BYTE_ORDER == BIG_ENDIAN) \
|| (var == UTF_16BE && BYTE_ORDER == LITTLE_ENDIAN)) \
data->__flags |= __GCONV_SWAP; \
} \
const int swap = data->__flags & __GCONV_SWAP;
#define EXTRA_LOOP_ARGS , swap
/* Direction of the transformation. */
enum direction
{
illegal_dir,
to_utf16,
from_utf16
};
enum variant
{
illegal_var,
UTF_16,
UTF_16LE,
UTF_16BE
};
struct utf16_data
{
enum direction dir;
enum variant var;
};
extern int gconv_init (struct __gconv_step *step);
int
gconv_init (struct __gconv_step *step)
{
/* Determine which direction. */
struct utf16_data *new_data;
enum direction dir = illegal_dir;
enum variant var = illegal_var;
int result;
if (__strcasecmp (step->__from_name, "UTF-16//") == 0)
{
dir = from_utf16;
var = UTF_16;
}
else if (__strcasecmp (step->__to_name, "UTF-16//") == 0)
{
dir = to_utf16;
var = UTF_16;
}
else if (__strcasecmp (step->__from_name, "UTF-16BE//") == 0)
{
dir = from_utf16;
var = UTF_16BE;
}
else if (__strcasecmp (step->__to_name, "UTF-16BE//") == 0)
{
dir = to_utf16;
var = UTF_16BE;
}
else if (__strcasecmp (step->__from_name, "UTF-16LE//") == 0)
{
dir = from_utf16;
var = UTF_16LE;
}
else if (__strcasecmp (step->__to_name, "UTF-16LE//") == 0)
{
dir = to_utf16;
var = UTF_16LE;
}
result = __GCONV_NOCONV;
if (__builtin_expect (dir, to_utf16) != illegal_dir)
{
new_data = (struct utf16_data *) malloc (sizeof (struct utf16_data));
result = __GCONV_NOMEM;
if (new_data != NULL)
{
new_data->dir = dir;
new_data->var = var;
step->__data = new_data;
if (dir == from_utf16)
{
step->__min_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;
}
else
{
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 = MAX_NEEDED_FROM;
}
step->__stateful = 0;
result = __GCONV_OK;
}
}
return result;
}
extern void gconv_end (struct __gconv_step *data);
void
gconv_end (struct __gconv_step *data)
{
free (data->__data);
}
/* Convert from the internal (UCS4-like) format to UTF-16. */
#define MIN_NEEDED_INPUT MIN_NEEDED_TO
#define MIN_NEEDED_OUTPUT MIN_NEEDED_FROM
#define MAX_NEEDED_OUTPUT MAX_NEEDED_FROM
#define LOOPFCT TO_LOOP
#define BODY \
{ \
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_mark_illegal_input (step_data); \
if (! ignore_errors_p ()) \
break; \
inptr += 4; \
++*irreversible; \
continue; \
} \
\
if (swap) \
{ \
if (__glibc_unlikely (c >= 0x10000)) \
{ \
if (__glibc_unlikely (c >= 0x110000)) \
{ \
STANDARD_TO_LOOP_ERR_HANDLER (4); \
} \
\
/* Generate a surrogate character. */ \
if (__glibc_unlikely (outptr + 4 > outend)) \
{ \
/* Overflow in the output buffer. */ \
result = __GCONV_FULL_OUTPUT; \
break; \
} \
\
put16 (outptr, bswap_16 (0xd7c0 + (c >> 10))); \
outptr += 2; \
put16 (outptr, bswap_16 (0xdc00 + (c & 0x3ff))); \
} \
else \
put16 (outptr, bswap_16 (c)); \
} \
else \
{ \
if (__glibc_unlikely (c >= 0x10000)) \
{ \
if (__glibc_unlikely (c >= 0x110000)) \
{ \
STANDARD_TO_LOOP_ERR_HANDLER (4); \
} \
\
/* Generate a surrogate character. */ \
if (__glibc_unlikely (outptr + 4 > outend)) \
{ \
/* Overflow in the output buffer. */ \
result = __GCONV_FULL_OUTPUT; \
break; \
} \
\
put16 (outptr, 0xd7c0 + (c >> 10)); \
outptr += 2; \
put16 (outptr, 0xdc00 + (c & 0x3ff)); \
} \
else \
put16 (outptr, c); \
} \
outptr += 2; \
inptr += 4; \
}
#define LOOP_NEED_FLAGS
#define EXTRA_LOOP_DECLS \
, int swap
#include <iconv/loop.c>
/* Convert from UTF-16 to the internal (UCS4-like) format. */
#define MIN_NEEDED_INPUT MIN_NEEDED_FROM
#define MAX_NEEDED_INPUT MAX_NEEDED_FROM
#define MIN_NEEDED_OUTPUT MIN_NEEDED_TO
#define LOOPFCT FROM_LOOP
#define BODY \
{ \
uint16_t u1 = get16 (inptr); \
\
if (swap) \
{ \
u1 = bswap_16 (u1); \
\
if (__builtin_expect (u1 < 0xd800, 1) || u1 > 0xdfff) \
{ \
/* No surrogate. */ \
put32 (outptr, u1); \
inptr += 2; \
} \
else \
{ \
uint16_t u2; \
\
if (__glibc_unlikely (u1 >= 0xdc00)) \
{ \
/* This is no valid first word for a surrogate. */ \
STANDARD_FROM_LOOP_ERR_HANDLER (2); \
} \
\
/* It's a surrogate character. At least the first word says \
it is. */ \
if (__glibc_unlikely (inptr + 4 > inend)) \
{ \
/* We don't have enough input for another complete input \
character. */ \
result = __GCONV_INCOMPLETE_INPUT; \
break; \
} \
\
inptr += 2; \
u2 = bswap_16 (get16 (inptr)); \
if (__builtin_expect (u2 < 0xdc00, 0) \
|| __builtin_expect (u2 > 0xdfff, 0)) \
{ \
/* This is no valid second word for a surrogate. */ \
inptr -= 2; \
STANDARD_FROM_LOOP_ERR_HANDLER (2); \
} \
\
put32 (outptr, ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00)); \
inptr += 2; \
} \
} \
else \
{ \
if (__builtin_expect (u1 < 0xd800, 1) || u1 > 0xdfff) \
{ \
/* No surrogate. */ \
put32 (outptr, u1); \
inptr += 2; \
} \
else \
{ \
if (__glibc_unlikely (u1 >= 0xdc00)) \
{ \
/* This is no valid first word for a surrogate. */ \
STANDARD_FROM_LOOP_ERR_HANDLER (2); \
} \
\
/* It's a surrogate character. At least the first word says \
it is. */ \
if (__glibc_unlikely (inptr + 4 > inend)) \
{ \
/* We don't have enough input for another complete input \
character. */ \
result = __GCONV_INCOMPLETE_INPUT; \
break; \
} \
\
inptr += 2; \
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. */ \
inptr -= 2; \
STANDARD_FROM_LOOP_ERR_HANDLER (2); \
} \
\
put32 (outptr, ((u1 - 0xd7c0) << 10) + (u2 - 0xdc00)); \
inptr += 2; \
} \
} \
outptr += 4; \
}
#define LOOP_NEED_FLAGS
#define EXTRA_LOOP_DECLS \
, int swap
#include <iconv/loop.c>
/* Now define the toplevel functions. */
#include <iconv/skeleton.c>
Reference in New Issue View Git Blame Copy Permalink