mirror of
https://sourceware.org/git/glibc.git
synced 2024-11-27 15:30:07 +00:00
d64b6ad075
* Versions.def: Add GLIBC_2.2 for libc. * iconv/gconv.h: Make header suitable for inclusion in public header by protecting all names with __. * iconv/gconv.c: Adapt for symbol name changes. * iconv/gconv.h: Likewise. * iconv/gconv_builtin.c: Likewise. * iconv/gconv_close.c: Likewise. * iconv/gconv_db.c: Likewise. * iconv/gconv_dl.c: Likewise. * iconv/gconv_int.h: Likewise. * iconv/gconv_open.c: Likewise. * iconv/gconv_simple.c: Likewise. * iconv/iconv.c: Likewise. * iconv/iconv_close.c: Likewise. * iconv/iconv_open.c: Likewise. * iconv/loop.c: Likewise. * iconv/skeleton.c: Likewise. * iconvdata/8bit-gap.c: Likewise. * iconvdata/8bit-generic.c: Likewise. * iconvdata/ansi_x3.110.c: Likewise. * iconvdata/big5.c: Likewise. * iconvdata/cns11643.h: Likewise. * iconvdata/cns11643l1.h: Likewise. * iconvdata/euc-cn.c: Likewise. * iconvdata/euc-jp.c: Likewise. * iconvdata/euc-kr.c: Likewise. * iconvdata/euc-tw.c: Likewise. * iconvdata/gb2312.h: Likewise. * iconvdata/iso-2022-jp.c: Likewise. * iconvdata/iso-2022-kr.c: Likewise. * iconvdata/iso646.c: Likewise. * iconvdata/iso8859-1.c: Likewise. * iconvdata/iso_6937-2.c: Likewise. * iconvdata/iso_6937.c: Likewise. * iconvdata/jis0201.h: Likewise. * iconvdata/jis0208.h: Likewise. * iconvdata/jis0212.h: Likewise. * iconvdata/johab.c: Likewise. * iconvdata/ksc5601.h: Likewise. * iconvdata/sjis.c: Likewise. * iconvdata/t.61.c: Likewise. * iconvdata/uhc.c: Likewise. * stdlib/mblen.c: Likewise. * stdlib/mbtowc.c: Likewise. * stdlib/wctomb.c: Likewise. * wcsmbs/btowc.c: Likewise. * wcsmbs/mbrtowc.c: Likewise. * wcsmbs/mbsnrtowcs.c: Likewise. * wcsmbs/mbsrtowcs.c: Likewise. * wcsmbs/wchar.h: Likewise. * wcsmbs/wcrtomb.c: Likewise. * wcsmbs/wcsmbsload.c: Likewise. * wcsmbs/wcsmbsload.h: Likewise. * wcsmbs/wcsnrtombs.c: Likewise. * wcsmbs/wcsrtombs.c: Likewise. * wcsmbs/wctob.c: Likewise. * include/limits.h (MB_LEN_MAX): Increase to 16. * sysdeps/generic/_G_config.h: Define _G_fpos_t as struct. Define _G_iconv_t. * sysdeps/unix/sysv/linux/_G_config.h: Likewise. * include/wchar.h: Change mbstate_t to __mbstate_t. * libio/Makefile (routines): Add wfiledoalloc, oldiofgetpos, oldiofgetpos64, oldiofsetpos, oldiofsetpos64, fputwc, fputwc_u, getwc, getwc_u, getwchar, getwchar_u, iofgetws, iofgetws_u, iofputws, iofputws_u, iogetwline, iowpadn, ioungetwc, putwc, putwc_u, putchar, putchar_u, swprintf, vwprintf, wprintf, wscanf, fwscanf, vwscanf, vswprintf, iovswscanf, swscanf, wgenops, wstrops, wfileops, and iofwide. (tests): Add tst_swprintf, tst_wprintf, tst_swscanf, and tst_wscanf. * libio/Versions: Add _IO_fgetpos, _IO_fgetpos64, _IO_fsetpos, _IO_fsetpos64, fgetpos, fgetpos64, fgetwc, fgetwc_unlocked, fgetws, fgetws_unlocked, fputwc, fputwc_unlocked, fputws, fputws_unlocked, fsetpos, fsetpos64, fwide, fwprintf, fwscanf, getwc, getwc_unlocked, getwchar, getwchar_unlocked, putwc, putwc_unlocked, putwchar, putwchar_unlocked, swprintf, swscanf, ungetwc, vfwprintf, vswprintf, vwprintf, vfwscanf, vswscanf, vwscanf, wprintf, and wscanf to GLIBC_2.2 for libc. * libio/libio.h: Define codecvt struct. Define _IO_wide_data. Extend _IO_file contain pointer to codecvt, widedata and mode. (_IO_getwc_unlocked): New macro. (_IO_putwc_unlocked): New macro. (_IO_fwide): New macro. * libio/libioP.h: Add new prototypes and adjust existing declarations. * libio/fileops.c (_IO_new_file_close_it): Reset normal or widedata buffers based on mode. (new_do_write): Set _IO_write_end to _IO_buf_end if stream is wide oriented. (_IO_new_file_overflow): Don't depend only on _IO_CURRENTLY_PUTTING flag to be enough to signal unallocated buffer. For wide oriented stream don't make it linebuffered. Don't use _IO_do_flush, use _IO_new_do_write directly. (_IO_new_file_seekoff): Change return value type to _IO_off64_t. (_IO_file_seek): Likewise. * libio/genops.c (_IO_least_marker): Make global. (__underflow): Orient stream if not already done. (__uflow): Likewise. (_IO_default_seekpos): Change to type _IO_off64_t. (_IO_default_seekoff): Likewise. (_IO_default_seek): Likewise. (_IO_no_init): New function. Similar to _IO_init but allows to orient in initialization. * libio/iolibio.h: Add prototype for _IO_vswprintf. Change _IO_pos_BAD to use _IO_off64_t. * libio/ftello.c: Use _IO_off_t. For now abort when use with wide char stream. * libio/ftello64.c: Likewise. * libio/ioftell.c: Likewise. * libio/iofopncook.c: Likewise. * libio/ioseekoff.c: Likewise. * libio/ioseekpos.c: Likewise. * libio/oldfileops.c: Likewise. * libio/iofgetpos.c: Store state of conversion if necessary. * libio/iofgetpos64.c: Likewise. * libio/iofsetpos.c: Restore conversion state if necessary. * libio/iofsetpos64.c: Likewise. * libio/iofdopen.c: Initialize so that stream can be wide oriented. * libio/iofopen.c: Likewise. * libio/iofopen64.c: Likewise. * libio/iopopen.c: Likewise. * libio/iovdprintf.c: Likewise. * libio/iovsprintf.c: Likewise. * libio/iovsscanf.c: Likewise. * libio/memstream.c: Likewise. * libio/obprintf.c: Likewise. * libio/iofputs.c: Orient stream if not already happened. * libio/iofputs_u.c: Likewise. * libio/iofwrite.c: Likewise. * libio/iofwrite_u.c: Likewise. * libio/ioputs.c: Likewise. * libio/iosetbuffer.c: Handle not yet oriented stream. * libio/iosetvbuf.c: Likewise. * libio/oldstdfiles.c: Adjust FILEBUF_LITERAL call. * libio/stdfiles.c: Likewise. * libio/strops.c (_IO_str_overflow): Correctly free buffer after failed allocation. (_IO_str_seekoff): Use _IO_off64_t. * libio/vasprintf.c: Pre-orient stream. * libio/vsnprintf.c: Likewise. * libio/fputwc.c: New file. * libio/fputwc_u.c: New file. * libio/fwprintf.c: New file. * libio/fwscanf.c: New file. * libio/getwc.c: New file. * libio/getwc_u.c: New file. * libio/getwchar.c: New file. * libio/getwchar_u.c: New file. * libio/iofgetws.c: New file. * libio/iofgetws_u.c: New file. * libio/iofputws.c: New file. * libio/iofputws_u.c: New file. * libio/iofwide.c: New file. * libio/iogetwline.c: New file. * libio/ioungetwc.c: New file. * libio/iovswscanf.c: New file. * libio/iowpadn.c: New file. * libio/oldiofgetpos.c: New file. * libio/oldiofgetpos64.c: New file. * libio/oldiofsetpos.c: New file. * libio/oldiofsetpos64.c: New file. * libio/putwc.c: New file. * libio/putwc_u.c: New file. * libio/putwchar.c: New file. * libio/putwchar_u.c: New file. * libio/swprintf.c: New file. * libio/swscanf.c: New file. * libio/tst_swprintf.c: New file. * libio/tst_swscanf.c: New file. * libio/tst_wprintf.c: New file. * libio/tst_wscanf.c: New file. * libio/tst_wscanf.input: New file. * libio/vswprintf.c: New file. * libio/vwprintf.c: New file. * libio/vwscanf.c: New file. * libio/wfiledoalloc.c: New file. * libio/wfileops.c: New file. * libio/wgenops.c: New file. * libio/wprintf.c: New file. * libio/wscanf.c: New file. * libio/wstrops.c: New file. * stdio-common/Makefile (routines): Add _itowa, itowa-digits, vfwprintf, and vfwscanf. * stdio-common/_itoa.c (base_table): Rename to _IO_base_table and make global. * stdio-common/_itowa.c: New file. * stdio-common/_itowa.h: New file. * stdio-common/itoa-digits.c: Minimal optimization. * stdio-common/itowa-digits.c: New file. * stdio-common/printf-parse.h: Allow use in wide character context. * stdio-common/printf-prs.c: Define ISASCII and MBRLEN. * stdio-common/printf.h (printf_info): Add wide bit. * stdio-common/printf_fp.c: Determine from wide bit whether stream is wide oriented or not. * stdio-common/printf_size.c: Likewise. * sysdeps/generic/printf_fphex.c: Likewise. * stdlib/strfmon.c: Call __printf_fp with wide bit cleared. * stdio-common/vfprintf.c: Rewrite to allow use in wide character context. * stdio-common/vfscand.c: Likewise. * stdio-common/vfwprintf.c: New file. * stdio-common/vfwscanf.c: New file. * time/Makefile (routines): Add wcsftime. (tests): Add tst_wcsftime. * time/Versions: Add wcsftime to GLIBC_2.2 for libc. * time/strftime.c: Make usable as wcsftime. * time/wcsftime.c: New file. * time/tst_wcsftime.c: New file. * wcsmbs/Makefile (routines): Add wmempcpy and wcschrnul. * wcsmbs/Versions: Add wmempcpy and wcschrnul to GLIBC_2.2 for libc. * wcsmbs/wcschrnul.c: New file. * wcsmbs/wmemcpy.c: New file. * wcsmbs/wmemcpy.c: Rename to __wmemcpy and make wmemcpy weak alias. * wcsmbs/wmemmove.c: Likewise for wmemmove. * manual/stdio.texi: Document is_char and wide element if printf_info. * manual/time.texi: Document wcsftime. * include/wchar.h: Add prototypes for __wmemcpy, __wmempcpy, __wmemmove, __wcschrnul, and __vfwscanf. * locale/langinfo.h: Add new LC_TIME entries for wchar_t data. * locale/C-time.c: Adapt for above change. * locale/categories.def: Likewise. * locale/localeinfo.h: Likewise. * localedata/Makefile: Don't run tests for now.
794 lines
25 KiB
C
794 lines
25 KiB
C
/* Conversion module for ISO-2022-JP.
|
|
Copyright (C) 1998, 1999 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. */
|
|
|
|
#include <gconv.h>
|
|
#include <stdint.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include "jis0201.h"
|
|
#include "jis0208.h"
|
|
#include "jis0212.h"
|
|
#include "gb2312.h"
|
|
#include "ksc5601.h"
|
|
|
|
struct gap
|
|
{
|
|
uint16_t start;
|
|
uint16_t end;
|
|
int32_t idx;
|
|
};
|
|
|
|
#include "iso8859-7jp.h"
|
|
|
|
/* This makes obvious what everybody knows: 0x1b is the Esc character. */
|
|
#define ESC 0x1b
|
|
|
|
/* We provide our own initialization and destructor function. */
|
|
#define DEFINE_INIT 0
|
|
#define DEFINE_FINI 0
|
|
|
|
/* Definitions used in the body of the `gconv' function. */
|
|
#define FROM_LOOP from_iso2022jp_loop
|
|
#define TO_LOOP to_iso2022jp_loop
|
|
#define MIN_NEEDED_FROM 1
|
|
#define MAX_NEEDED_FROM 4
|
|
#define MIN_NEEDED_TO 4
|
|
#define MAX_NEEDED_TO 4
|
|
#define FROM_DIRECTION (dir == from_iso2022jp)
|
|
#define PREPARE_LOOP \
|
|
enum direction dir = ((struct iso2022jp_data *) step->__data)->dir; \
|
|
enum variant var = ((struct iso2022jp_data *) step->__data)->var; \
|
|
int save_set; \
|
|
int *setp = &data->__statep->count;
|
|
#define EXTRA_LOOP_ARGS , var, setp
|
|
|
|
|
|
/* Direction of the transformation. */
|
|
enum direction
|
|
{
|
|
illegal_dir,
|
|
to_iso2022jp,
|
|
from_iso2022jp
|
|
};
|
|
|
|
/* We handle ISO-2022-jp and ISO-2022-JP-2 here. */
|
|
enum variant
|
|
{
|
|
illegal_var,
|
|
iso2022jp,
|
|
iso2022jp2
|
|
};
|
|
|
|
|
|
struct iso2022jp_data
|
|
{
|
|
enum direction dir;
|
|
enum variant var;
|
|
};
|
|
|
|
|
|
/* The COUNT element of the state keeps track of the currently selected
|
|
character set. The possible values are: */
|
|
enum
|
|
{
|
|
ASCII_set = 0,
|
|
JISX0208_1978_set,
|
|
JISX0208_1983_set,
|
|
JISX0201_Roman_set,
|
|
JISX0201_Kana_set,
|
|
GB2312_set,
|
|
KSC5601_set,
|
|
JISX0212_set
|
|
};
|
|
|
|
/* The second value stored is the designation of the G2 set. The following
|
|
values are possible: */
|
|
enum
|
|
{
|
|
UNSPECIFIED_set = 0,
|
|
ISO88591_set,
|
|
ISO88597_set
|
|
};
|
|
|
|
|
|
int
|
|
gconv_init (struct __gconv_step *step)
|
|
{
|
|
/* Determine which direction. */
|
|
struct iso2022jp_data *new_data;
|
|
enum direction dir = illegal_dir;
|
|
enum variant var = illegal_var;
|
|
int result;
|
|
|
|
if (__strcasecmp (step->__from_name, "ISO-2022-JP//") == 0)
|
|
{
|
|
dir = from_iso2022jp;
|
|
var = iso2022jp;
|
|
}
|
|
else if (__strcasecmp (step->__to_name, "ISO-2022-JP//") == 0)
|
|
{
|
|
dir = to_iso2022jp;
|
|
var = iso2022jp;
|
|
}
|
|
else if (__strcasecmp (step->__from_name, "ISO-2022-JP-2//") == 0)
|
|
{
|
|
dir = from_iso2022jp;
|
|
var = iso2022jp2;
|
|
}
|
|
else if (__strcasecmp (step->__to_name, "ISO-2022-JP-2//") == 0)
|
|
{
|
|
dir = to_iso2022jp;
|
|
var = iso2022jp2;
|
|
}
|
|
|
|
result = __GCONV_NOCONV;
|
|
if (dir != illegal_dir)
|
|
{
|
|
new_data
|
|
= (struct iso2022jp_data *) malloc (sizeof (struct iso2022jp_data));
|
|
|
|
result = __GCONV_NOMEM;
|
|
if (new_data != NULL)
|
|
{
|
|
new_data->dir = dir;
|
|
new_data->var = var;
|
|
step->__data = new_data;
|
|
|
|
if (dir == from_iso2022jp)
|
|
{
|
|
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
|
|
{
|
|
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 + 2;
|
|
}
|
|
|
|
/* Yes, this is a stateful encoding. */
|
|
step->__stateful = 1;
|
|
|
|
result = __GCONV_OK;
|
|
}
|
|
}
|
|
|
|
return result;
|
|
}
|
|
|
|
|
|
void
|
|
gconv_end (struct __gconv_step *data)
|
|
{
|
|
free (data->__data);
|
|
}
|
|
|
|
|
|
/* Since this is a stateful encoding we have to provide code which resets
|
|
the output state to the initial state. This has to be done during the
|
|
flushing. */
|
|
#define EMIT_SHIFT_TO_INIT \
|
|
if (data->__statep->count != ASCII_set) \
|
|
{ \
|
|
enum direction dir = ((struct iso2022jp_data *) step->__data)->dir; \
|
|
\
|
|
if (dir == from_iso2022jp) \
|
|
/* It's easy, we don't have to emit anything, we just reset the \
|
|
state for the input. Note that this also clears the G2 \
|
|
designation. */ \
|
|
data->__statep->count = ASCII_set; \
|
|
else \
|
|
{ \
|
|
unsigned char *outbuf = data->__outbuf; \
|
|
\
|
|
/* We are not in the initial state. To switch back we have \
|
|
to emit the sequence `Esc ( B'. */ \
|
|
if (outbuf + 3 > data->__outbufend) \
|
|
/* We don't have enough room in the output buffer. */ \
|
|
status = __GCONV_FULL_OUTPUT; \
|
|
else \
|
|
{ \
|
|
/* Write out the shift sequence. */ \
|
|
*outbuf++ = ESC; \
|
|
*outbuf++ = '('; \
|
|
*outbuf++ = 'B'; \
|
|
data->__outbuf = outbuf; \
|
|
/* Note that this also clears the G2 designation. */ \
|
|
data->__statep->count = ASCII_set; \
|
|
} \
|
|
} \
|
|
}
|
|
|
|
|
|
/* Since we might have to reset input pointer we must be able to save
|
|
and retore the state. */
|
|
#define SAVE_RESET_STATE(Save) \
|
|
if (Save) \
|
|
save_set = *setp; \
|
|
else \
|
|
*setp = save_set
|
|
|
|
|
|
/* First define the conversion function from ISO-2022-JP to UCS4. */
|
|
#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 \
|
|
{ \
|
|
uint32_t ch = *inptr; \
|
|
\
|
|
/* Recognize escape sequences. */ \
|
|
if (ch == ESC) \
|
|
{ \
|
|
/* We now must be prepared to read two to three more \
|
|
chracters. If we have a match in the first character but \
|
|
then the input buffer ends we terminate with an error since \
|
|
we must not risk missing an escape sequence just because it \
|
|
is not entirely in the current input buffer. */ \
|
|
if (inptr + 2 >= inend \
|
|
|| (var == iso2022jp2 && inptr[1] == '$' && inptr[2] == '(' \
|
|
&& inptr + 3 >= inend)) \
|
|
{ \
|
|
/* Not enough input available. */ \
|
|
result = __GCONV_EMPTY_INPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
if (inptr[1] == '(') \
|
|
{ \
|
|
if (inptr[2] == 'B') \
|
|
{ \
|
|
/* ASCII selected. */ \
|
|
set = ASCII_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
else if (inptr[2] == 'J') \
|
|
{ \
|
|
/* JIS X 0201 selected. */ \
|
|
set = JISX0201_Roman_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
else if (var == iso2022jp2 && inptr[2] == 'I') \
|
|
{ \
|
|
/* JIS X 0201 selected. */ \
|
|
set = JISX0201_Kana_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
} \
|
|
else if (inptr[1] == '$') \
|
|
{ \
|
|
if (inptr[2] == '@') \
|
|
{ \
|
|
/* JIS X 0208-1978 selected. */ \
|
|
set = JISX0208_1978_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
else if (inptr[2] == 'B') \
|
|
{ \
|
|
/* JIS X 0208-1983 selected. */ \
|
|
set = JISX0208_1983_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
else if (var == iso2022jp2) \
|
|
{ \
|
|
if (inptr[2] == 'A') \
|
|
{ \
|
|
/* GB 2312-1980 selected. */ \
|
|
set = GB2312_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
else if (inptr[2] == '(') \
|
|
{ \
|
|
if (inptr[3] == 'C') \
|
|
{ \
|
|
/* KSC 5601-1987 selected. */ \
|
|
set = KSC5601_set; \
|
|
inptr += 4; \
|
|
continue; \
|
|
} \
|
|
else if (inptr[3] == 'D') \
|
|
{ \
|
|
/* JIS X 0212-1990 selected. */ \
|
|
set = JISX0212_set; \
|
|
inptr += 4; \
|
|
continue; \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
else if (var == iso2022jp2 && inptr[1] == '.') \
|
|
{ \
|
|
if (inptr[2] == 'A') \
|
|
{ \
|
|
/* ISO 8859-1-GR selected. */ \
|
|
set2 = ISO88591_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
else if (inptr[2] == 'F') \
|
|
{ \
|
|
/* ISO 8859-7-GR selected. */ \
|
|
set2 = ISO88597_set; \
|
|
inptr += 3; \
|
|
continue; \
|
|
} \
|
|
} \
|
|
} \
|
|
\
|
|
if (ch == ESC && var == iso2022jp2 && inptr[1] == 'N') \
|
|
{ \
|
|
if (set2 == ISO88591_set) \
|
|
{ \
|
|
ch = inptr[2] | 0x80; \
|
|
inptr += 3; \
|
|
} \
|
|
else if (set2 == ISO88597_set) \
|
|
{ \
|
|
/* We use the table from the ISO 8859-7 module. */ \
|
|
if (inptr[2] < 0x20 || inptr[2] > 0x80) \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
ch = iso88597_to_ucs4[inptr[2] - 0x20]; \
|
|
if (ch == 0) \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
inptr += 3; \
|
|
} \
|
|
else \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
} \
|
|
else if (set == ASCII_set || (ch < 0x21 || ch == 0x7f)) \
|
|
/* Almost done, just advance the input pointer. */ \
|
|
++inptr; \
|
|
else if (set == JISX0201_Roman_set) \
|
|
{ \
|
|
/* Use the JIS X 0201 table. */ \
|
|
ch = jisx0201_to_ucs4 (ch); \
|
|
if (ch == __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
++inptr; \
|
|
} \
|
|
else if (set == JISX0201_Kana_set) \
|
|
{ \
|
|
/* Use the JIS X 0201 table. */ \
|
|
ch = jisx0201_to_ucs4 (ch + 0x80); \
|
|
if (ch == __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
++inptr; \
|
|
} \
|
|
else \
|
|
{ \
|
|
if (set == JISX0208_1978_set || set == JISX0208_1983_set) \
|
|
/* XXX I don't have the tables for these two old variants of \
|
|
JIS X 0208. Therefore I'm using the tables for JIS X \
|
|
0208-1990. If somebody has problems with this please \
|
|
provide the appropriate tables. */ \
|
|
ch = jisx0208_to_ucs4 (&inptr, \
|
|
NEED_LENGTH_TEST ? inend - inptr : 2, 0); \
|
|
else if (set == JISX0212_set) \
|
|
/* Use the JIS X 0212 table. */ \
|
|
ch = jisx0212_to_ucs4 (&inptr, \
|
|
NEED_LENGTH_TEST ? inend - inptr : 2, 0); \
|
|
else if (set == GB2312_set) \
|
|
/* Use the GB 2312 table. */ \
|
|
ch = gb2312_to_ucs4 (&inptr, \
|
|
NEED_LENGTH_TEST ? inend - inptr : 2, 0); \
|
|
else \
|
|
{ \
|
|
assert (set == KSC5601_set); \
|
|
\
|
|
/* Use the KSC 5601 table. */ \
|
|
ch = ksc5601_to_ucs4 (&inptr, \
|
|
NEED_LENGTH_TEST ? inend - inptr : 2, 0); \
|
|
} \
|
|
\
|
|
if (NEED_LENGTH_TEST && ch == 0) \
|
|
{ \
|
|
result = __GCONV_EMPTY_INPUT; \
|
|
break; \
|
|
} \
|
|
else if (ch == __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
} \
|
|
\
|
|
*((uint32_t *) outptr)++ = ch; \
|
|
}
|
|
#define EXTRA_LOOP_DECLS , enum variant var, int *setp
|
|
#define INIT_PARAMS int set = *setp % 0x100, set2 = *setp / 0x100
|
|
#define UPDATE_PARAMS *setp = (set2 << 8) + set
|
|
#include <iconv/loop.c>
|
|
|
|
|
|
/* Next, define the other direction. */
|
|
#define MIN_NEEDED_INPUT MIN_NEEDED_TO
|
|
#define MIN_NEEDED_OUTPUT MIN_NEEDED_FROM
|
|
#define MAX_NEEDED_OUTPUT (MAX_NEEDED_FROM + 2)
|
|
#define LOOPFCT TO_LOOP
|
|
#define BODY \
|
|
{ \
|
|
uint32_t ch; \
|
|
size_t written = 0; \
|
|
\
|
|
ch = *((uint32_t *) inptr); \
|
|
\
|
|
/* First see whether we can write the character using the currently \
|
|
selected character set. */ \
|
|
if (set == ASCII_set) \
|
|
{ \
|
|
/* Please note that the NUL byte is *not* matched if we are not \
|
|
currently using the ASCII charset. This is because we must \
|
|
switch to the initial state whenever a NUL byte is written. */ \
|
|
if (ch <= 0x7f) \
|
|
{ \
|
|
*outptr++ = ch; \
|
|
written = 1; \
|
|
} \
|
|
/* At the beginning of a line, G2 designation is cleared. */ \
|
|
if (var == iso2022jp2 && ch == 0x0a) \
|
|
set2 = UNSPECIFIED_set; \
|
|
} \
|
|
else if (set == JISX0201_Roman_set) \
|
|
{ \
|
|
unsigned char buf[2]; \
|
|
written = ucs4_to_jisx0201 (ch, buf); \
|
|
if (written != __UNKNOWN_10646_CHAR && buf[0] > 0x20 \
|
|
&& buf[0] < 0x80) \
|
|
{ \
|
|
*outptr++ = buf[0]; \
|
|
written = 1; \
|
|
} \
|
|
else \
|
|
written = __UNKNOWN_10646_CHAR; \
|
|
} \
|
|
else if (set == JISX0201_Kana_set) \
|
|
{ \
|
|
unsigned char buf[2]; \
|
|
written = ucs4_to_jisx0201 (ch, buf); \
|
|
if (written != __UNKNOWN_10646_CHAR && buf[0] > 0xa0 \
|
|
&& buf[0] < 0xe0) \
|
|
{ \
|
|
*outptr++ = buf[0] - 0x80; \
|
|
written = 1; \
|
|
} \
|
|
else \
|
|
written = __UNKNOWN_10646_CHAR; \
|
|
} \
|
|
else \
|
|
{ \
|
|
if (set == JISX0208_1978_set || set == JISX0208_1983_set) \
|
|
written = ucs4_to_jisx0208 (ch, outptr, \
|
|
(NEED_LENGTH_TEST \
|
|
? outend - outptr : 2)); \
|
|
else if (set == JISX0212_set) \
|
|
written = ucs4_to_jisx0212 (ch, outptr, \
|
|
(NEED_LENGTH_TEST \
|
|
? outend - outptr : 2)); \
|
|
else if (set == GB2312_set) \
|
|
written = ucs4_to_gb2312 (ch, outptr, (NEED_LENGTH_TEST \
|
|
? outend - outptr : 2)); \
|
|
else \
|
|
{ \
|
|
assert (set == KSC5601_set); \
|
|
\
|
|
written = ucs4_to_ksc5601 (ch, outptr, \
|
|
(NEED_LENGTH_TEST \
|
|
? outend - outptr : 2)); \
|
|
} \
|
|
\
|
|
if (NEED_LENGTH_TEST && written == 0) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
else if (written != __UNKNOWN_10646_CHAR) \
|
|
outptr += written; \
|
|
} \
|
|
\
|
|
if (written == __UNKNOWN_10646_CHAR || written == 0) \
|
|
{ \
|
|
if (set2 == ISO88591_set) \
|
|
{ \
|
|
if (ch >= 0x80 && ch <= 0xff) \
|
|
{ \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = 'N'; \
|
|
*outptr++ = ch & 0x7f; \
|
|
written = 3; \
|
|
} \
|
|
} \
|
|
else if (set2 == ISO88597_set) \
|
|
{ \
|
|
const struct gap *rp = from_idx; \
|
|
\
|
|
while (ch > rp->end) \
|
|
++rp; \
|
|
if (ch >= rp->start) \
|
|
{ \
|
|
unsigned char res = iso88597_from_ucs4[ch - 0xa0 + rp->idx]; \
|
|
if (res != '\0') \
|
|
{ \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = 'N'; \
|
|
*outptr++ = res; \
|
|
written = 3; \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
\
|
|
if (written == __UNKNOWN_10646_CHAR || written == 0) \
|
|
{ \
|
|
/* Either this is an unknown character or we have to switch \
|
|
the currently selected character set. The character sets \
|
|
do not code entirely separate parts of ISO 10646 and \
|
|
therefore there is no single correct result. If we choose \
|
|
the character set to use wrong we might be end up with \
|
|
using yet another character set for the next character \
|
|
though the current and the next could be encoded with one \
|
|
character set. We leave this kind of optimization for \
|
|
later and now simply use a fixed order in which we test for \
|
|
availability */ \
|
|
\
|
|
if (ch <= 0x7f) \
|
|
{ \
|
|
/* We must encode using ASCII. First write out the \
|
|
escape sequence. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 4 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '('; \
|
|
*outptr++ = 'B'; \
|
|
set = ASCII_set; \
|
|
*outptr++ = ch; \
|
|
\
|
|
/* At the beginning of a line, G2 designation is cleared. */ \
|
|
if (var == iso2022jp2 && ch == 0x0a) \
|
|
set2 = UNSPECIFIED_set; \
|
|
} \
|
|
else \
|
|
{ \
|
|
/* Now it becomes difficult. We must search the other \
|
|
character sets one by one and we cannot use simple \
|
|
arithmetic to determine whether the character can be \
|
|
encoded using this set. */ \
|
|
size_t written; \
|
|
unsigned char buf[2]; \
|
|
\
|
|
written = ucs4_to_jisx0201 (ch, buf); \
|
|
if (written != __UNKNOWN_10646_CHAR && buf[0] < 0x80) \
|
|
{ \
|
|
/* We use JIS X 0201. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 4 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '('; \
|
|
*outptr++ = 'J'; \
|
|
set = JISX0201_Roman_set; \
|
|
*outptr++ = buf[0]; \
|
|
} \
|
|
else \
|
|
{ \
|
|
written = ucs4_to_jisx0208 (ch, buf, 2); \
|
|
if (written != __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
/* We use JIS X 0208. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 5 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '$'; \
|
|
*outptr++ = 'B'; \
|
|
set = JISX0208_1983_set; \
|
|
*outptr++ = buf[0]; \
|
|
*outptr++ = buf[1]; \
|
|
} \
|
|
else if (var == iso2022jp) \
|
|
{ \
|
|
/* We have no other choice. */ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
else \
|
|
{ \
|
|
written = ucs4_to_jisx0212 (ch, buf, 2); \
|
|
if (written != __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
/* We use JIS X 0212. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 6 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '$'; \
|
|
*outptr++ = '('; \
|
|
*outptr++ = 'D'; \
|
|
set = JISX0212_set; \
|
|
*outptr++ = buf[0]; \
|
|
*outptr++ = buf[1]; \
|
|
} \
|
|
else \
|
|
{ \
|
|
written = ucs4_to_jisx0201 (ch, buf); \
|
|
if (written != __UNKNOWN_10646_CHAR \
|
|
&& buf[0] >= 0x80) \
|
|
{ \
|
|
/* We use JIS X 0201. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 4 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '('; \
|
|
*outptr++ = 'I'; \
|
|
set = JISX0201_Kana_set; \
|
|
*outptr++ = buf[0] - 0x80; \
|
|
} \
|
|
else if (ch != 0xa5 && ch >= 0x80 && ch <= 0xff) \
|
|
{ \
|
|
/* ISO 8859-1 upper half. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 6 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '.'; \
|
|
*outptr++ = 'A'; \
|
|
set2 = ISO88591_set; \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = 'N'; \
|
|
*outptr++ = ch; \
|
|
} \
|
|
else \
|
|
{ \
|
|
written = ucs4_to_gb2312 (ch, buf, 2); \
|
|
if (written != __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
/* We use GB 2312. */ \
|
|
if (NEED_LENGTH_TEST && outptr + 5 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
\
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '$'; \
|
|
*outptr++ = 'A'; \
|
|
set = GB2312_set; \
|
|
*outptr++ = buf[0]; \
|
|
*outptr++ = buf[1]; \
|
|
} \
|
|
else \
|
|
{ \
|
|
written = ucs4_to_ksc5601 (ch, buf, 2); \
|
|
if (written != __UNKNOWN_10646_CHAR) \
|
|
{ \
|
|
/* We use KSC 5601. */ \
|
|
if (NEED_LENGTH_TEST \
|
|
&& outptr + 6 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '$'; \
|
|
*outptr++ = '('; \
|
|
*outptr++ = 'C'; \
|
|
set = KSC5601_set; \
|
|
*outptr++ = buf[0]; \
|
|
*outptr++ = buf[1]; \
|
|
} \
|
|
else \
|
|
{ \
|
|
const struct gap *rp = from_idx; \
|
|
unsigned char gch = 0; \
|
|
\
|
|
while (ch > rp->end) \
|
|
++rp; \
|
|
if (ch >= rp->start) \
|
|
{ \
|
|
ch = ch - 0xa0 + rp->idx; \
|
|
gch = iso88597_from_ucs4[ch]; \
|
|
} \
|
|
\
|
|
if (gch != 0) \
|
|
{ \
|
|
/* We use ISO 8859-7 greek. */ \
|
|
if (NEED_LENGTH_TEST \
|
|
&& outptr + 6 > outend) \
|
|
{ \
|
|
result = __GCONV_FULL_OUTPUT; \
|
|
break; \
|
|
} \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = '.'; \
|
|
*outptr++ = 'F'; \
|
|
set2 = ISO88597_set; \
|
|
*outptr++ = ESC; \
|
|
*outptr++ = 'N'; \
|
|
*outptr++ = gch; \
|
|
} \
|
|
else \
|
|
{ \
|
|
result = __GCONV_ILLEGAL_INPUT; \
|
|
break; \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
} \
|
|
\
|
|
/* Now that we wrote the output increment the input pointer. */ \
|
|
inptr += 4; \
|
|
}
|
|
#define EXTRA_LOOP_DECLS , enum variant var, int *setp
|
|
#define INIT_PARAMS int set = *setp % 0x100, set2 = *setp / 0x100
|
|
#define UPDATE_PARAMS *setp = (set2 << 8) + set
|
|
#include <iconv/loop.c>
|
|
|
|
|
|
/* Now define the toplevel functions. */
|
|
#include <iconv/skeleton.c>
|