mirror of
https://sourceware.org/git/glibc.git
synced 2024-12-29 05:51:10 +00:00
978ce92b89
2000-04-20 Ulrich Drepper <drepper@redhat.com> * iconv/skeleton.c: Add some more __builtin_expect. * iconv/loop.c: Likewise.
554 lines
17 KiB
C
554 lines
17 KiB
C
/* Skeleton for a conversion module.
|
|
Copyright (C) 1998, 1999, 2000 Free Software Foundation, Inc.
|
|
This file is part of the GNU C Library.
|
|
Contributed by Ulrich Drepper <drepper@cygnus.com>, 1998.
|
|
|
|
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.
|
|
|
|
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.
|
|
|
|
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. */
|
|
|
|
/* This file can be included to provide definitions of several things
|
|
many modules have in common. It can be customized using the following
|
|
macros:
|
|
|
|
DEFINE_INIT define the default initializer. This requires the
|
|
following symbol to be defined.
|
|
|
|
CHARSET_NAME string with official name of the coded character
|
|
set (in all-caps)
|
|
|
|
DEFINE_FINI define the default destructor function.
|
|
|
|
MIN_NEEDED_FROM minimal number of bytes needed for the from-charset.
|
|
MIN_NEEDED_TO likewise for the to-charset.
|
|
|
|
MAX_NEEDED_FROM maximal number of bytes needed for the from-charset.
|
|
This macro is optional, it defaults to MIN_NEEDED_FROM.
|
|
MAX_NEEDED_TO likewise for the to-charset.
|
|
|
|
DEFINE_DIRECTION_OBJECTS
|
|
two objects will be defined to be used when the
|
|
`gconv' function must only distinguish two
|
|
directions. This is implied by DEFINE_INIT.
|
|
If this macro is not defined the following
|
|
macro must be available.
|
|
|
|
FROM_DIRECTION this macro is supposed to return a value != 0
|
|
if we convert from the current character set,
|
|
otherwise it return 0.
|
|
|
|
EMIT_SHIFT_TO_INIT this symbol is optional. If it is defined it
|
|
defines some code which writes out a sequence
|
|
of characters which bring the current state into
|
|
the initial state.
|
|
|
|
FROM_LOOP name of the function implementing the conversion
|
|
from the current characters.
|
|
TO_LOOP likewise for the other direction
|
|
|
|
RESET_STATE in case of an error we must reset the state for
|
|
the rerun so this macro must be defined for
|
|
stateful encodings. It takes an argument which
|
|
is nonzero when saving.
|
|
|
|
RESET_INPUT_BUFFER If the input character sets allow this the macro
|
|
can be defined to reset the input buffer pointers
|
|
to cover only those characters up to the error.
|
|
|
|
FUNCTION_NAME if not set the conversion function is named `gconv'.
|
|
|
|
PREPARE_LOOP optional code preparing the conversion loop. Can
|
|
contain variable definitions.
|
|
END_LOOP also optional, may be used to store information
|
|
|
|
EXTRA_LOOP_ARGS optional macro specifying extra arguments passed
|
|
to loop function.
|
|
*/
|
|
|
|
#include <assert.h>
|
|
#include <gconv.h>
|
|
#include <string.h>
|
|
#define __need_size_t
|
|
#define __need_NULL
|
|
#include <stddef.h>
|
|
|
|
#ifndef STATIC_GCONV
|
|
# include <dlfcn.h>
|
|
#endif
|
|
|
|
#ifndef DL_CALL_FCT
|
|
# define DL_CALL_FCT(fct, args) fct args
|
|
#endif
|
|
|
|
/* The direction objects. */
|
|
#if DEFINE_DIRECTION_OBJECTS || DEFINE_INIT
|
|
static int from_object;
|
|
static int to_object;
|
|
|
|
# ifndef FROM_DIRECTION
|
|
# define FROM_DIRECTION (step->__data == &from_object)
|
|
# endif
|
|
#else
|
|
# ifndef FROM_DIRECTION
|
|
# error "FROM_DIRECTION must be provided if direction objects are not used"
|
|
# endif
|
|
#endif
|
|
|
|
|
|
/* How many bytes are needed at most for the from-charset. */
|
|
#ifndef MAX_NEEDED_FROM
|
|
# define MAX_NEEDED_FROM MIN_NEEDED_FROM
|
|
#endif
|
|
|
|
/* Same for the to-charset. */
|
|
#ifndef MAX_NEEDED_TO
|
|
# define MAX_NEEDED_TO MIN_NEEDED_TO
|
|
#endif
|
|
|
|
|
|
/* Define macros which can access unaligned buffers. These macros are
|
|
supposed to be used only in code outside the inner loops. For the inner
|
|
loops we have other definitions which allow optimized access. */
|
|
#ifdef _STRING_ARCH_unaligned
|
|
/* We can handle unaligned memory access. */
|
|
# define get16u(addr) *((uint16_t *) (addr))
|
|
# define get32u(addr) *((uint32_t *) (addr))
|
|
|
|
/* We need no special support for writing values either. */
|
|
# define put16u(addr, val) *((uint16_t *) (addr)) = (val)
|
|
# define put32u(addr, val) *((uint32_t *) (addr)) = (val)
|
|
#else
|
|
/* Distinguish between big endian and little endian. */
|
|
# if __BYTE_ORDER == __LITTLE_ENDIAN
|
|
# define get16u(addr) \
|
|
(((__const unsigned char *) (addr))[1] << 8 \
|
|
| ((__const unsigned char *) (addr))[0])
|
|
# define get32u(addr) \
|
|
(((((__const unsigned char *) (addr))[3] << 8 \
|
|
| ((__const unsigned char *) (addr))[2]) << 8 \
|
|
| ((__const unsigned char *) (addr))[1]) << 8 \
|
|
| ((__const unsigned char *) (addr))[0])
|
|
|
|
# define put16u(addr, val) \
|
|
({ uint16_t __val = (val); \
|
|
((unsigned char *) (addr))[0] = __val; \
|
|
((unsigned char *) (addr))[1] = __val >> 8; \
|
|
(void) 0; })
|
|
# define put32u(addr, val) \
|
|
({ uint32_t __val = (val); \
|
|
((unsigned char *) (addr))[0] = __val; \
|
|
__val >>= 8; \
|
|
((unsigned char *) (addr))[1] = __val; \
|
|
__val >>= 8; \
|
|
((unsigned char *) (addr))[2] = __val; \
|
|
__val >>= 8; \
|
|
((unsigned char *) (addr))[3] = __val; \
|
|
(void) 0; })
|
|
# else
|
|
# define get16u(addr) \
|
|
(((__const unsigned char *) (addr))[0] << 8 \
|
|
| ((__const unsigned char *) (addr))[1])
|
|
# define get32u(addr) \
|
|
(((((__const unsigned char *) (addr))[0] << 8 \
|
|
| ((__const unsigned char *) (addr))[1]) << 8 \
|
|
| ((__const unsigned char *) (addr))[2]) << 8 \
|
|
| ((__const unsigned char *) (addr))[3])
|
|
|
|
# define put16u(addr, val) \
|
|
({ uint16_t __val = (val); \
|
|
((unsigned char *) (addr))[1] = __val; \
|
|
((unsigned char *) (addr))[0] = __val >> 8; \
|
|
(void) 0; })
|
|
# define put32u(addr, val) \
|
|
({ uint32_t __val = (val); \
|
|
((unsigned char *) (addr))[3] = __val; \
|
|
__val >>= 8; \
|
|
((unsigned char *) (addr))[2] = __val; \
|
|
__val >>= 8; \
|
|
((unsigned char *) (addr))[1] = __val; \
|
|
__val >>= 8; \
|
|
((unsigned char *) (addr))[0] = __val; \
|
|
(void) 0; })
|
|
# endif
|
|
#endif
|
|
|
|
|
|
/* For conversions from a fixed width character sets to another fixed width
|
|
character set we we can define RESET_INPUT_BUFFER is necessary. */
|
|
#if !defined RESET_INPUT_BUFFER && !defined SAVE_RESET_STATE
|
|
# if MIN_NEEDED_FROM == MAX_NEEDED_FROM && MIN_NEEDED_TO == MAX_NEEDED_TO
|
|
/* We have to use these `if's here since the compiler cannot know that
|
|
(outbuf - outerr) is always divisible by MIN_NEEDED_TO. */
|
|
# define RESET_INPUT_BUFFER \
|
|
if (MIN_NEEDED_FROM % MIN_NEEDED_TO == 0) \
|
|
*inptrp -= (outbuf - outerr) * (MIN_NEEDED_FROM / MIN_NEEDED_TO); \
|
|
else if (MIN_NEEDED_TO % MIN_NEEDED_FROM == 0) \
|
|
*inptrp -= (outbuf - outerr) / (MIN_NEEDED_TO / MIN_NEEDED_FROM); \
|
|
else \
|
|
*inptrp -= ((outbuf - outerr) / MIN_NEEDED_TO) * MIN_NEEDED_FROM
|
|
# endif
|
|
#endif
|
|
|
|
|
|
/* The default init function. It simply matches the name and initializes
|
|
the step data to point to one of the objects above. */
|
|
#if DEFINE_INIT
|
|
# ifndef CHARSET_NAME
|
|
# error "CHARSET_NAME not defined"
|
|
# endif
|
|
|
|
int
|
|
gconv_init (struct __gconv_step *step)
|
|
{
|
|
/* Determine which direction. */
|
|
if (strcmp (step->__from_name, CHARSET_NAME) == 0)
|
|
{
|
|
step->__data = &from_object;
|
|
|
|
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 = MAX_NEEDED_TO;
|
|
}
|
|
else if (strcmp (step->__to_name, CHARSET_NAME) == 0)
|
|
{
|
|
step->__data = &to_object;
|
|
|
|
step->__min_needed_from = MIN_NEEDED_TO;
|
|
step->__max_needed_from = MAX_NEEDED_TO;
|
|
step->__min_needed_to = MIN_NEEDED_FROM;
|
|
step->__max_needed_to = MAX_NEEDED_FROM;
|
|
}
|
|
else
|
|
return __GCONV_NOCONV;
|
|
|
|
#ifdef RESET_STATE
|
|
step->__stateful = 1;
|
|
#else
|
|
step->__stateful = 0;
|
|
#endif
|
|
|
|
return __GCONV_OK;
|
|
}
|
|
#endif
|
|
|
|
|
|
/* The default destructor function does nothing in the moment and so
|
|
be define it at all. But we still provide the macro just in case
|
|
we need it some day. */
|
|
#if DEFINE_FINI
|
|
#endif
|
|
|
|
|
|
/* If no arguments have to passed to the loop function define the macro
|
|
as empty. */
|
|
#ifndef EXTRA_LOOP_ARGS
|
|
# define EXTRA_LOOP_ARGS
|
|
#endif
|
|
|
|
|
|
/* This is the actual conversion function. */
|
|
#ifndef FUNCTION_NAME
|
|
# define FUNCTION_NAME gconv
|
|
#endif
|
|
|
|
/* The macros are used to access the function to convert single characters. */
|
|
#define SINGLE(fct) SINGLE2 (fct)
|
|
#define SINGLE2(fct) fct##_single
|
|
|
|
|
|
int
|
|
FUNCTION_NAME (struct __gconv_step *step, struct __gconv_step_data *data,
|
|
const unsigned char **inptrp, const unsigned char *inend,
|
|
size_t *written, int do_flush, int consume_incomplete)
|
|
{
|
|
struct __gconv_step *next_step = step + 1;
|
|
struct __gconv_step_data *next_data = data + 1;
|
|
__gconv_fct fct = data->__is_last ? NULL : next_step->__fct;
|
|
int status;
|
|
|
|
/* If the function is called with no input this means we have to reset
|
|
to the initial state. The possibly partly converted input is
|
|
dropped. */
|
|
if (__builtin_expect (do_flush, 0))
|
|
{
|
|
status = __GCONV_OK;
|
|
|
|
#ifdef EMIT_SHIFT_TO_INIT
|
|
/* Emit the escape sequence to reset the state. */
|
|
EMIT_SHIFT_TO_INIT;
|
|
#endif
|
|
/* Call the steps down the chain if there are any but only if we
|
|
successfully emitted the escape sequence. */
|
|
if (status == __GCONV_OK && ! data->__is_last)
|
|
status = DL_CALL_FCT (fct, (next_step, next_data, NULL, NULL,
|
|
written, 1, consume_incomplete));
|
|
}
|
|
else
|
|
{
|
|
/* We preserve the initial values of the pointer variables. */
|
|
const unsigned char *inptr = *inptrp;
|
|
unsigned char *outbuf = data->__outbuf;
|
|
unsigned char *outend = data->__outbufend;
|
|
unsigned char *outstart;
|
|
/* This variable is used to count the number of characters we
|
|
actually converted. */
|
|
size_t converted = 0;
|
|
#if defined _STRING_ARCH_unaligned \
|
|
|| MIN_NEEDED_FROM == 1 || MAX_NEEDED_FROM % MIN_NEEDED_FROM != 0 \
|
|
|| MIN_NEEDED_TO == 1 || MAX_NEEDED_TO % MIN_NEEDED_TO != 0
|
|
# define unaligned 0
|
|
#else
|
|
int unaligned;
|
|
# define GEN_unaligned(name) GEN_unaligned2 (name)
|
|
# define GEN_unaligned2(name) name##_unaligned
|
|
#endif
|
|
|
|
#ifdef PREPARE_LOOP
|
|
PREPARE_LOOP
|
|
#endif
|
|
|
|
#if MAX_NEEDED_FROM > 1 || MAX_NEEDED_TO > 1
|
|
/* If the function is used to implement the mb*towc*() or wc*tomb*()
|
|
functions we must test whether any bytes from the last call are
|
|
stored in the `state' object. */
|
|
if (((MAX_NEEDED_FROM > 1 && FROM_DIRECTION)
|
|
|| (MAX_NEEDED_TO > 1 && !FROM_DIRECTION))
|
|
&& consume_incomplete && (data->__statep->__count & 7) != 0)
|
|
{
|
|
/* Yep, we have some bytes left over. Process them now. */
|
|
|
|
# if MAX_NEEDED_FROM > 1
|
|
if (MAX_NEEDED_TO == 1 || FROM_DIRECTION)
|
|
status = SINGLE(FROM_LOOP) (inptrp, inend, &outbuf, outend,
|
|
data->__statep, step->__data,
|
|
&converted EXTRA_LOOP_ARGS);
|
|
# endif
|
|
# if MAX_NEEDED_FROM > 1 && MAX_NEEDED_TO > 1 && !ONE_DIRECTION
|
|
else
|
|
# endif
|
|
# if MAX_NEEDED_TO > 1 && !ONE_DIRECTION
|
|
status = SINGLE(TO_LOOP) (inptrp, inend, &outbuf, outend,
|
|
data->__statep, step->__data,
|
|
&converted EXTRA_LOOP_ARGS);
|
|
# endif
|
|
|
|
if (status != __GCONV_OK)
|
|
return status;
|
|
}
|
|
#endif
|
|
|
|
#if !defined _STRING_ARCH_unaligned \
|
|
&& MIN_NEEDED_FROM != 1 && MAX_NEEDED_FROM % MIN_NEEDED_FROM == 0 \
|
|
&& MIN_NEEDED_TO != 1 && MAX_NEEDED_TO % MIN_NEEDED_TO == 0
|
|
/* The following assumes that encodings, which have a variable length
|
|
what might unalign a buffer even though it is a aligned in the
|
|
beginning, either don't have the minimal number of bytes as a divisor
|
|
of the maximum length or have a minimum length of 1. This is true
|
|
for all known and supported encodings. */
|
|
unaligned = ((FROM_DIRECTION
|
|
&& ((uintptr_t) inptr % MIN_NEEDED_FROM != 0
|
|
|| (data->__is_last
|
|
&& (uintptr_t) outbuf % MIN_NEEDED_TO != 0)))
|
|
|| (!FROM_DIRECTION
|
|
&& ((data->__is_last
|
|
&& (uintptr_t) outbuf % MIN_NEEDED_FROM != 0)
|
|
|| (uintptr_t) inptr % MIN_NEEDED_TO != 0)));
|
|
#endif
|
|
|
|
do
|
|
{
|
|
/* Remember the start value for this round. */
|
|
inptr = *inptrp;
|
|
/* The outbuf buffer is empty. */
|
|
outstart = outbuf;
|
|
|
|
#ifdef SAVE_RESET_STATE
|
|
SAVE_RESET_STATE (1);
|
|
#endif
|
|
|
|
if (!unaligned)
|
|
{
|
|
if (FROM_DIRECTION)
|
|
/* Run the conversion loop. */
|
|
status = FROM_LOOP (inptrp, inend, &outbuf, outend,
|
|
data->__statep, step->__data, &converted
|
|
EXTRA_LOOP_ARGS);
|
|
else
|
|
/* Run the conversion loop. */
|
|
status = TO_LOOP (inptrp, inend, &outbuf, outend,
|
|
data->__statep, step->__data, &converted
|
|
EXTRA_LOOP_ARGS);
|
|
}
|
|
#if !defined _STRING_ARCH_unaligned \
|
|
&& MIN_NEEDED_FROM != 1 && MAX_NEEDED_FROM % MIN_NEEDED_FROM == 0 \
|
|
&& MIN_NEEDED_TO != 1 && MAX_NEEDED_TO % MIN_NEEDED_TO == 0
|
|
else
|
|
{
|
|
if (FROM_DIRECTION)
|
|
/* Run the conversion loop. */
|
|
status = GEN_unaligned (FROM_LOOP) (inptrp, inend, &outbuf,
|
|
outend, data->__statep,
|
|
step->__data, &converted
|
|
EXTRA_LOOP_ARGS);
|
|
else
|
|
/* Run the conversion loop. */
|
|
status = GEN_unaligned (TO_LOOP) (inptrp, inend, &outbuf,
|
|
outend, data->__statep,
|
|
step->__data, &converted
|
|
EXTRA_LOOP_ARGS);
|
|
}
|
|
#endif
|
|
|
|
/* We finished one use of the loops. */
|
|
++data->__invocation_counter;
|
|
|
|
/* If this is the last step leave the loop, there is nothing
|
|
we can do. */
|
|
if (data->__is_last)
|
|
{
|
|
/* Store information about how many bytes are available. */
|
|
data->__outbuf = outbuf;
|
|
|
|
/* Remember how many non-identical characters we converted. */
|
|
*written += converted;
|
|
|
|
break;
|
|
}
|
|
|
|
/* Write out all output which was produced. */
|
|
if (outbuf > outstart)
|
|
{
|
|
const unsigned char *outerr = data->__outbuf;
|
|
int result;
|
|
|
|
result = DL_CALL_FCT (fct, (next_step, next_data, &outerr,
|
|
outbuf, written, 0,
|
|
consume_incomplete));
|
|
|
|
if (result != __GCONV_EMPTY_INPUT)
|
|
{
|
|
if (__builtin_expect (outerr != outbuf, 0))
|
|
{
|
|
#ifdef RESET_INPUT_BUFFER
|
|
RESET_INPUT_BUFFER;
|
|
#else
|
|
/* We have a problem with the in on of the functions
|
|
below. Undo the conversion upto the error point. */
|
|
size_t nstatus;
|
|
|
|
/* Reload the pointers. */
|
|
*inptrp = inptr;
|
|
outbuf = outstart;
|
|
|
|
/* Reset the state. */
|
|
# ifdef SAVE_RESET_STATE
|
|
SAVE_RESET_STATE (0);
|
|
# endif
|
|
|
|
if (FROM_DIRECTION)
|
|
/* Run the conversion loop. */
|
|
nstatus = FROM_LOOP ((const unsigned char **) inptrp,
|
|
(const unsigned char *) inend,
|
|
(unsigned char **) &outbuf,
|
|
(unsigned char *) outerr,
|
|
data->__statep, step->__data,
|
|
&converted EXTRA_LOOP_ARGS);
|
|
else
|
|
/* Run the conversion loop. */
|
|
nstatus = TO_LOOP ((const unsigned char **) inptrp,
|
|
(const unsigned char *) inend,
|
|
(unsigned char **) &outbuf,
|
|
(unsigned char *) outerr,
|
|
data->__statep, step->__data,
|
|
&converted EXTRA_LOOP_ARGS);
|
|
|
|
/* We must run out of output buffer space in this
|
|
rerun. */
|
|
assert (outbuf == outerr);
|
|
assert (nstatus == __GCONV_FULL_OUTPUT);
|
|
|
|
/* If we haven't consumed a single byte decrement
|
|
the invocation counter. */
|
|
if (outbuf == outstart)
|
|
--data->__invocation_counter;
|
|
#endif /* reset input buffer */
|
|
}
|
|
|
|
/* Change the status. */
|
|
status = result;
|
|
}
|
|
else
|
|
/* All the output is consumed, we can make another run
|
|
if everything was ok. */
|
|
if (status == __GCONV_FULL_OUTPUT)
|
|
status = __GCONV_OK;
|
|
}
|
|
}
|
|
while (status == __GCONV_OK);
|
|
|
|
#ifdef END_LOOP
|
|
END_LOOP
|
|
#endif
|
|
|
|
/* If we are supposed to consume all character store now all of the
|
|
remaining characters in the `state' object. */
|
|
#if MAX_NEEDED_FROM > 1 || MAX_NEEDED_TO > 1
|
|
if (((MAX_NEEDED_FROM > 1 && FROM_DIRECTION)
|
|
|| (MAX_NEEDED_TO > 1 && !FROM_DIRECTION))
|
|
&& consume_incomplete && status == __GCONV_INCOMPLETE_INPUT)
|
|
{
|
|
# ifdef STORE_REST
|
|
mbstate_t *state = data->__statep;
|
|
|
|
STORE_REST
|
|
# else
|
|
size_t cnt;
|
|
|
|
/* Make sure the remaining bytes fit into the state objects
|
|
buffer. */
|
|
assert (inend - *inptrp < 4);
|
|
|
|
for (cnt = 0; *inptrp < inend; ++cnt)
|
|
data->__statep->__value.__wchb[cnt] = *(*inptrp)++;
|
|
data->__statep->__count &= ~7;
|
|
data->__statep->__count |= cnt;
|
|
# endif
|
|
}
|
|
#endif
|
|
}
|
|
|
|
return status;
|
|
}
|
|
|
|
#undef DEFINE_INIT
|
|
#undef CHARSET_NAME
|
|
#undef DEFINE_FINI
|
|
#undef MIN_NEEDED_FROM
|
|
#undef MIN_NEEDED_TO
|
|
#undef MAX_NEEDED_FROM
|
|
#undef MAX_NEEDED_TO
|
|
#undef DEFINE_DIRECTION_OBJECTS
|
|
#undef FROM_DIRECTION
|
|
#undef EMIT_SHIFT_TO_INIT
|
|
#undef FROM_LOOP
|
|
#undef TO_LOOP
|
|
#undef RESET_STATE
|
|
#undef RESET_INPUT_BUFFER
|
|
#undef FUNCTION_NAME
|
|
#undef PREPARE_LOOP
|
|
#undef END_LOOP
|
|
#undef ONE_DIRECTION
|
|
#undef STORE_REST
|