Update.
1998-04-25 18:39 Ulrich Drepper <drepper@cygnus.com> * iconvdata/Makefile: Use gap method for iso8859-5, iso8859-7, iso8859-8 and iso8859-10. * iconvdata/iso8859-5.c: Change to use gap method. * iconvdata/iso8859-7.c: Likewise. * iconvdata/iso8859-8.c: Likewise. * iconvdata/iso8859-10.c: Likewise. * iconvdata/Makefile: Add rules for ISO-2022-JP module. * iconv/skeleton.c: Allow END_LOOP do be defined and use it at the end of the loop. * iconvdata/iso-2022-jp.c: New file. * iconvdata/ksc5601.c: Don't use uint16_t to represent byte sequence. * iconvdata/ksc5601.h: Unify function interfaces. * iconvdata/euckr.c: Adapt for changed ksc5601.h interface. * iconvdata/uhc.c: Likewise. * iconvdata/gb2312.h: Use correct types. * iconvdata/iso646.c (gconv_open): Correctly initialize the character size elements of data.
This commit is contained in:
Ulrich Drepper
parent
05c9fa3c05
commit
918b9d72a9
24
ChangeLog
24
ChangeLog
|
|
@ -1,3 +1,27 @@
|
|||
1998-04-25 18:39 Ulrich Drepper <drepper@cygnus.com>
|
||||
|
||||
* iconvdata/Makefile: Use gap method for iso8859-5, iso8859-7,
|
||||
iso8859-8 and iso8859-10.
|
||||
* iconvdata/iso8859-5.c: Change to use gap method.
|
||||
* iconvdata/iso8859-7.c: Likewise.
|
||||
* iconvdata/iso8859-8.c: Likewise.
|
||||
* iconvdata/iso8859-10.c: Likewise.
|
||||
|
||||
* iconvdata/Makefile: Add rules for ISO-2022-JP module.
|
||||
* iconv/skeleton.c: Allow END_LOOP do be defined and use it at the
|
||||
end of the loop.
|
||||
* iconvdata/iso-2022-jp.c: New file.
|
||||
|
||||
* iconvdata/ksc5601.c: Don't use uint16_t to represent byte sequence.
|
||||
* iconvdata/ksc5601.h: Unify function interfaces.
|
||||
* iconvdata/euckr.c: Adapt for changed ksc5601.h interface.
|
||||
* iconvdata/uhc.c: Likewise.
|
||||
|
||||
* iconvdata/gb2312.h: Use correct types.
|
||||
|
||||
* iconvdata/iso646.c (gconv_open): Correctly initialize the character
|
||||
size elements of data.
|
||||
|
||||
1998-04-24 Ulrich Drepper <drepper@cygnus.com>
|
||||
|
||||
* string/tst-svc.expect: Adapt for change in .input.
|
||||
|
|
|
|||
|
|
@ -70,6 +70,7 @@
|
|||
|
||||
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.
|
||||
|
|
@ -144,10 +145,20 @@ gconv_init (struct gconv_step *step)
|
|||
else
|
||||
return GCONV_NOCONV;
|
||||
|
||||
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;
|
||||
if (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
|
||||
{
|
||||
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;
|
||||
}
|
||||
|
||||
#ifdef RESET_STATE
|
||||
step->stateful = 1;
|
||||
|
|
@ -334,6 +345,10 @@ FUNCTION_NAME (struct gconv_step *step, struct gconv_step_data *data,
|
|||
|
||||
/* Remember how many characters we converted. */
|
||||
*written += converted;
|
||||
|
||||
#ifdef END_LOOP
|
||||
END_LOOP
|
||||
#endif
|
||||
}
|
||||
|
||||
return status;
|
||||
|
|
@ -354,3 +369,5 @@ FUNCTION_NAME (struct gconv_step *step, struct gconv_step_data *data,
|
|||
#undef RESET_STATE
|
||||
#undef RESET_INPUT_BUFFER
|
||||
#undef FUNCTION_NAME
|
||||
#undef PREPARE_LOOP
|
||||
#undef END_LOOP
|
||||
|
|
|
|||
|
|
@ -38,7 +38,7 @@ modules += KOI8-R LATIN-GREEK LATIN-GREEK-1 IBM256 IBM273 IBM277 IBM278 \
|
|||
IBM850 IBM851 IBM852 IBM855 IBM857 IBM860 IBM861 IBM862 \
|
||||
IBM863 IBM864 IBM865 IBM868 IBM869 IBM875 IBM880 IBM918 \
|
||||
IBM1004 IBM1026 CP1250 CP1251 CP1252 CP1253 CP1254 CP1255 \
|
||||
CP1256 CP1257
|
||||
CP1256 CP1257 ISO-2022-JP
|
||||
endif
|
||||
modules.so := $(addsuffix .so, $(modules))
|
||||
|
||||
|
|
@ -142,6 +142,7 @@ BIG5-routines := big5
|
|||
EUC-JP-routines := eucjp
|
||||
EUC-CN-routines := euccn
|
||||
EUC-TW-routines := euctw
|
||||
ISO-2022-JP-routines := iso-2022-jp
|
||||
libJIS-routines := jis0201 jis0208 jis0212
|
||||
libKSC-routines := ksc5601
|
||||
libGB-routines := gb2312
|
||||
|
|
@ -160,6 +161,10 @@ $(objpfx)EUC-CN.so: $(objpfx)libGB.so
|
|||
LDFLAGS-EUC-TW.so = -Wl,-rpath,$(gconvdir)
|
||||
$(objpfx)EUC-TW.so: $(objpfx)libCNS.so
|
||||
|
||||
LDFLAGS-ISO-2022-JP.so = -Wl,-rpath,$(gconvdir)
|
||||
$(objpfx)ISO-2022-JP.so: $(objpfx)libJIS.so $(objpfx)libGB.so \
|
||||
$(objpfx)libCNS.so
|
||||
|
||||
LDFLAGS-libJIS.so = -Wl,-soname,$(@F)
|
||||
LDFLAGS-libKSC.so = -Wl,-soname,$(@F)
|
||||
LDFLAGS-libGB.so = -Wl,-soname,$(@F)
|
||||
|
|
@ -188,7 +193,7 @@ distribute := 8bit-generic.c 8bit-gap.c gap.pl gaptab.pl gconv-modules \
|
|||
ibm903.c ibm904.c ibm905.c ibm918.c ibm1004.c ibm1026.c \
|
||||
ibm1047.c cp1250.c cp1251.c cp1252.c cp1253.c cp1254.c \
|
||||
cp1255.c cp1256.c cp1257.c cp874.c cp874.h cp737.c cp737.h \
|
||||
cp775.c cp775.h
|
||||
cp775.c cp775.h iso-2022-jp.c
|
||||
|
||||
# We build the transformation modules only when we build shared libs.
|
||||
ifeq (yes,$(build-shared))
|
||||
|
|
@ -226,9 +231,8 @@ $(make-target-directory)
|
|||
echo "};" ) > $@.new && rm -f $@ && mv $@.new $@
|
||||
endef
|
||||
|
||||
sed-generated-headers := iso8859-2.h iso8859-3.h iso8859-4.h iso8859-5.h \
|
||||
iso8859-6.h iso8859-7.h iso8859-8.h iso8859-9.h \
|
||||
iso8859-10.h koi-8.h hp-roman8.h ebcdic-at-de.h \
|
||||
sed-generated-headers := iso8859-2.h iso8859-3.h iso8859-4.h iso8859-6.h \
|
||||
iso8859-9.h koi-8.h hp-roman8.h ebcdic-at-de.h \
|
||||
ebcdic-at-de-a.h ebcdic-ca-fr.h ebcdic-dk-no.h \
|
||||
ebcdic-dk-no-a.h ebcdic-es.h ebcdic-es-a.h \
|
||||
ebcdic-es-s.h ebcdic-fi-se.h ebcdic-fi-se-a.h \
|
||||
|
|
@ -236,7 +240,7 @@ sed-generated-headers := iso8859-2.h iso8859-3.h iso8859-4.h iso8859-5.h \
|
|||
ebcdic-pt.h ebcdic-uk.h ebcdic-us.h ibm037.h \
|
||||
ibm038.h ibm274.h ibm275.h ibm423.h ibm424.h \
|
||||
ibm500.h ibm870.h ibm871.h ibm891.h ibm903.h \
|
||||
ibm904.h ibm905.h ibm1047.h
|
||||
ibm904.h ibm905.h ibm1047.h iso8859-7jp.h
|
||||
|
||||
define generate-8bit-gap-table
|
||||
$(make-target-directory)
|
||||
|
|
@ -260,7 +264,9 @@ perl-generated-headers := koi8-r.h latin-greek.h latin-greek-1.h \
|
|||
ibm863.h ibm864.h ibm865.h ibm868.h ibm869.h \
|
||||
ibm875.h ibm880.h ibm918.h ibm1004.h ibm1026.h \
|
||||
cp1250.h cp1251.h cp1252.h cp1253.h cp1254.h \
|
||||
cp1255.h cp1256.h cp1257.h
|
||||
cp1255.h cp1256.h cp1257.h iso8859-5.h \
|
||||
iso8859-7.h iso8859-8.h iso8859-10.h \
|
||||
iso8859-7jp.h
|
||||
|
||||
# The headers must be generated before the compilation.
|
||||
before-compile = $(addprefix $(objpfx),$(sed-generated-headers)) \
|
||||
|
|
@ -274,18 +280,10 @@ $(objpfx)iso8859-3.h: ../localedata/charmaps/ISO-8859-3 Makefile
|
|||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-4.h: ../localedata/charmaps/ISO-8859-4 Makefile
|
||||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-5.h: ../localedata/charmaps/ISO-8859-5 Makefile
|
||||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-6.h: ../localedata/charmaps/ISO-8859-6 Makefile
|
||||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-7.h: ../localedata/charmaps/ISO-8859-7 Makefile
|
||||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-8.h: ../localedata/charmaps/ISO-8859-8 Makefile
|
||||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-9.h: ../localedata/charmaps/ISO-8859-9 Makefile
|
||||
$(generate-8bit-table)
|
||||
$(objpfx)iso8859-10.h: ../localedata/charmaps/ISO-8859-10 Makefile
|
||||
$(generate-8bit-table)
|
||||
|
||||
$(objpfx)koi-8.h: ../localedata/charmaps/KOI-8 Makefile
|
||||
$(generate-8bit-table)
|
||||
|
|
@ -352,7 +350,30 @@ $(objpfx)ibm905.h: ../localedata/charmaps/IBM905 Makefile
|
|||
$(objpfx)ibm1047.h: ../localedata/charmaps/IBM1047 Makefile
|
||||
$(generate-8bit-table)
|
||||
|
||||
$(objpfx)iso8859-7jp.h: ../localedata/charmaps/ISO-8859-7 Makefile
|
||||
$(make-target-directory)
|
||||
( echo "static const uint32_t iso88597_to_ucs4[96] = {"; \
|
||||
sed -e '/^[^[:space:]]*[[:space:]]*.x00/d' -e 's/^[^[:space:]]*[[:space:]]*.x\([A-F].\)[[:space:]]*<U\(....\)>.*/ [0x\1-0xA0] = 0x\2,/p' -e d $^ | sort -u; \
|
||||
echo "};"; \
|
||||
echo "static struct gap from_idx[] = {"; \
|
||||
sed -e '/^[^[:space:]]*[[:space:]]*.x00/d' -e 's/^[^[:space:]]*[[:space:]]*.x\([A-F].\)[[:space:]]*<U\(....\)>.*/0x\2 0x\1/p' -e d $^ | sort -u | $(PERL) gap.pl; \
|
||||
echo " { start: 0xffff, end: 0xffff, idx: 0 }"; \
|
||||
echo "};"; \
|
||||
echo "static const char iso88597_from_ucs4[] = {"; \
|
||||
sed -e '/^[^[:space:]]*[[:space:]]*.x00/d' -e 's/^[^[:space:]]*[[:space:]]*.x\([A-F].\)[[:space:]]*<U\(....\)>.*/0x\2 0x\1/p' -e d $^ | sort -u | $(PERL) gaptab.pl; \
|
||||
echo "};" ) > $@.new && rm -f $@ && mv $@.new $@
|
||||
|
||||
|
||||
ifneq ($(PERL),no)
|
||||
$(objpfx)iso8859-5.h: ../localedata/charmaps/ISO-8859-5 Makefile
|
||||
$(generate-8bit-gap-table)
|
||||
$(objpfx)iso8859-7.h: ../localedata/charmaps/ISO-8859-7 Makefile
|
||||
$(generate-8bit-gap-table)
|
||||
$(objpfx)iso8859-8.h: ../localedata/charmaps/ISO-8859-8 Makefile
|
||||
$(generate-8bit-gap-table)
|
||||
$(objpfx)iso8859-10.h: ../localedata/charmaps/ISO-8859-10 Makefile
|
||||
$(generate-8bit-gap-table)
|
||||
|
||||
$(objpfx)koi8-r.h: ../localedata/charmaps/KOI8-R Makefile
|
||||
$(generate-8bit-gap-table)
|
||||
$(objpfx)latin-greek.h: ../localedata/charmaps/LATIN-GREEK Makefile
|
||||
|
|
|
|||
|
|
@ -88,27 +88,20 @@ euckr_from_ucs4 (uint32_t ch, unsigned char *cp)
|
|||
{ \
|
||||
/* Two-byte character. First test whether the next character \
|
||||
is also available. */ \
|
||||
int ch2; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && inptr + 1 >= inend) \
|
||||
ch = ksc5601_to_ucs4 (&inptr, \
|
||||
NEED_LENGTH_TEST ? inptr - inbufend : 2, x080); \
|
||||
if (NEED_LENGTH_TEST && ch == 0) \
|
||||
{ \
|
||||
/* The second character is not available. */ \
|
||||
result = GCONV_INCOMPLETE_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
ch2 = inptr[1]; \
|
||||
\
|
||||
if (ch2 < 0xa1 || ch2 >= 0xfe \
|
||||
|| ((ch = ksc5601_to_ucs4 ((uint16_t) (ch * 256 + ch2) & 0x7f7f)) \
|
||||
== UNKNOWN_10646_CHAR)) \
|
||||
if (ch == UNKNOWN_10646_CHAR)) \
|
||||
{ \
|
||||
/* This is an illegal character. */ \
|
||||
result = GCONV_ILLEGAL_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
inptr += 2; \
|
||||
} \
|
||||
\
|
||||
*((uint32_t *) outptr)++ = ch; \
|
||||
|
|
|
|||
|
|
@ -29,7 +29,7 @@ extern const uint16_t __gb2312_to_ucs[];
|
|||
|
||||
|
||||
static inline uint32_t
|
||||
gb2312_to_ucs4 (const char **s, size_t avail, unsigned char offset)
|
||||
gb2312_to_ucs4 (const unsigned char **s, size_t avail, unsigned char offset)
|
||||
{
|
||||
unsigned char ch = *(*s);
|
||||
unsigned char ch2;
|
||||
|
|
@ -66,7 +66,7 @@ extern const char __gb2312_from_ucs4_tab8[][2];
|
|||
extern const char __gb2312_from_ucs4_tab9[][2];
|
||||
|
||||
static inline size_t
|
||||
ucs4_to_gb2312 (uint32_t wch, char *s, size_t avail)
|
||||
ucs4_to_gb2312 (uint32_t wch, unsigned char **s, size_t avail)
|
||||
{
|
||||
unsigned int ch = (unsigned int) wch;
|
||||
char buf[2];
|
||||
|
|
@ -212,11 +212,16 @@ ucs4_to_gb2312 (uint32_t wch, char *s, size_t avail)
|
|||
return UNKNOWN_10646_CHAR;
|
||||
}
|
||||
|
||||
if (cp[1] != '\0' && avail < 2)
|
||||
if (cp[0] == '\0')
|
||||
return UNKNOWN_10646_CHAR;
|
||||
|
||||
assert (cp[1] != '\0');
|
||||
|
||||
if (avail < 2)
|
||||
return 0;
|
||||
|
||||
s[0] = cp[0];
|
||||
s[1] = cp[1];
|
||||
*(*s)++ = cp[0];
|
||||
*(*s)++ = cp[1];
|
||||
|
||||
return 2;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -0,0 +1,693 @@
|
|||
/* Conversion module for ISO-2022-JP.
|
||||
Copyright (C) 1998 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 <string.h>
|
||||
#include "jis0201.h"
|
||||
#include "jis0212.h"
|
||||
#include "jis0201.h"
|
||||
#include "gb2312.h"
|
||||
#include "ksc5601.h"
|
||||
#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
|
||||
#define TO_LOOP to_iso2022jp
|
||||
#define MIN_NEEDED_FROM 1
|
||||
#define MAX_NEEDED_FROM 4
|
||||
#define MIN_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 set = data->statep->count;
|
||||
#define END_LOOP \
|
||||
data->statep->count = set;
|
||||
#define EXTRA_LOOP_ARGS , var, set
|
||||
|
||||
|
||||
/* 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;
|
||||
mbstate_t save_state;
|
||||
};
|
||||
|
||||
|
||||
/* 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_set,
|
||||
GB2312_set,
|
||||
KSC5601_set,
|
||||
JISX0212_set,
|
||||
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;
|
||||
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)))
|
||||
!= 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 = MIN_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 = MIN_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 != 0) \
|
||||
{ \
|
||||
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. */ \
|
||||
data->statep->count = 0; \
|
||||
else \
|
||||
{ \
|
||||
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; \
|
||||
data->statep->count = 0; \
|
||||
} \
|
||||
} \
|
||||
}
|
||||
|
||||
|
||||
/* 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) \
|
||||
((struct iso2022jp_data *) step->data)->save_state.count \
|
||||
= data->statep->count; \
|
||||
else \
|
||||
data->statep->count \
|
||||
= ((struct iso2022jp_data *) step->data)->save_state.count
|
||||
|
||||
|
||||
/* 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; \
|
||||
\
|
||||
/* This is a 7bit character set, disallow all 8bit characters. */ \
|
||||
if (ch > 0x7f) \
|
||||
{ \
|
||||
result = GCONV_ILLEGAL_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
/* 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 >= inbufend \
|
||||
|| (var == iso2022jp2 && inptr[1] == '$' && inptr[2] == '(' \
|
||||
&& inptr +3 >= inbufend)) \
|
||||
{ \
|
||||
/* 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_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 (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. */ \
|
||||
set = ISO88591_set; \
|
||||
inptr += 3; \
|
||||
continue; \
|
||||
} \
|
||||
else if (inptr[2] == 'F') \
|
||||
{ \
|
||||
/* ISO 8859-7-GR selected. */ \
|
||||
set = ISO88597_set; \
|
||||
inptr += 3; \
|
||||
continue; \
|
||||
} \
|
||||
} \
|
||||
} \
|
||||
\
|
||||
if (set == ASCII_set \
|
||||
|| (var < ISO88591_set && (ch < 0x21 || ch == 0x7f)) \
|
||||
|| (var >= ISO88591_set && ch < 0x20)) \
|
||||
/* Almost done, just advance the input pointer. */ \
|
||||
++inptr; \
|
||||
else if (set == JISX0201_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 == ISO88591_set) \
|
||||
{ \
|
||||
/* This is quite easy. All characters are defined and the \
|
||||
ISO 10646 value is computed by adding 0x80. */ \
|
||||
ch += 0x80; \
|
||||
++inptr; \
|
||||
} \
|
||||
else if (set == ISO88597_set) \
|
||||
{ \
|
||||
/* We use the table from the ISO 8859-7 module. */ \
|
||||
ch = iso88597_to_ucs4[ch - 0x20]; \
|
||||
if (ch == 0) \
|
||||
{ \
|
||||
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 ? inbufend - inptr : 2, 0); \
|
||||
else if (set == JISX0212_set) \
|
||||
/* Use the JIS X 0212 table. */ \
|
||||
ch = jisx0212_to_ucs4 (&inptr, \
|
||||
NEED_LENGTH_TEST ? inbufend - inptr : 2, 0); \
|
||||
else if (set == GB2312_set) \
|
||||
/* Use the GB 2312 table. */ \
|
||||
ch = gb2312_to_ucs4 (&inptr, \
|
||||
NEED_LENGTH_TEST ? inbufend - inptr : 2, 0); \
|
||||
else \
|
||||
{ \
|
||||
assert (set == KSC5601_set); \
|
||||
\
|
||||
/* Use the KSC 5601 table. */ \
|
||||
ch = ksc5601_to_ucs4 (&inptr, \
|
||||
NEED_LENGTH_TEST ? inbufend - 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 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 \
|
||||
{ \
|
||||
unsigned char 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 \
|
||||
|| (ch >= 0x01 && ((set < ISO88591_set && (ch < 0x21 || ch == 0x7f)) \
|
||||
|| (set >= ISO88591_set && ch < 0x20)))) \
|
||||
{ \
|
||||
/* 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; \
|
||||
} \
|
||||
} \
|
||||
else if (set == JISX0201_set) \
|
||||
written = ucs4_to_jisx0201 (ch, outptr); \
|
||||
else if (set == ISO88591_set) \
|
||||
{ \
|
||||
if (ch >= 0xa0 && ch <= 0xff) \
|
||||
{ \
|
||||
*outptr++ = ch - 0x80; \
|
||||
written = 1; \
|
||||
} \
|
||||
} \
|
||||
else if (set == ISO88597_set) \
|
||||
{ \
|
||||
const struct gap *rp = from_idx; \
|
||||
\
|
||||
while (ch > rp->end) \
|
||||
++rp; \
|
||||
if (ch >= rp->start) \
|
||||
{ \
|
||||
unsigned char res = iso88597_from_ucs4[ch + rp->idx]; \
|
||||
if (res != '\0') \
|
||||
{ \
|
||||
*outptr++ = res; \
|
||||
written = 1; \
|
||||
} \
|
||||
} \
|
||||
} \
|
||||
else \
|
||||
{ \
|
||||
if (set == JISX0208_1978_set || set == JISX0208_1983_set) \
|
||||
written = ucs4_to_jisx0208 (ch, outptr, \
|
||||
(NEED_LENGTH_TEST \
|
||||
? outbufend - outptr : 2)); \
|
||||
else if (set == JISX0212_set) \
|
||||
written = ucs4_to_jisx0212 (ch, outptr, \
|
||||
(NEED_LENGTH_TEST \
|
||||
? outbufend - outptr : 2)); \
|
||||
else if (set == GB2312_set) \
|
||||
written = ucs4_to_gb2312 (ch, outptr, (NEED_LENGTH_TEST \
|
||||
? outbufend - outptr : 2)); \
|
||||
else \
|
||||
{ \
|
||||
assert (set == KSC5601_set); \
|
||||
\
|
||||
written = ucs4_to_ksc5601 (ch, outptr, \
|
||||
(NEED_LENGTH_TEST \
|
||||
? outbufend - outptr : 2)); \
|
||||
} \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && written == 0) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
} \
|
||||
\
|
||||
if (written == UNKNOWN_10646_CHAR) \
|
||||
{ \
|
||||
/* 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 */ \
|
||||
\
|
||||
/* First test whether we have at least three more bytes for \
|
||||
the escape sequence. The two charsets which require four \
|
||||
bytes will be handled later. */ \
|
||||
if (NEED_LENGTH_TEST && outptr + 3 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
if (ch <= 0x7f) \
|
||||
{ \
|
||||
/* We must encode using ASCII. First write out the \
|
||||
escape sequence. */ \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '('; \
|
||||
*outptr++ = 'B'; \
|
||||
set = ASCII_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && outptr == outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = ch; \
|
||||
} \
|
||||
else if (ch >= 0xa0 && ch <= 0xff) \
|
||||
{ \
|
||||
/* This character set is not available in ISO-2022-JP. */ \
|
||||
if (var == iso2022jp) \
|
||||
{ \
|
||||
result == GCONV_ILLEGAL_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
/* We must use the ISO 8859-1 upper half. */ \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '.'; \
|
||||
*outptr++ = 'A'; \
|
||||
set = ISO88591_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && outptr == outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = ch - 0x80; \
|
||||
} \
|
||||
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) \
|
||||
{ \
|
||||
/* We use JIS X 0201. */ \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '$'; \
|
||||
*outptr++ = '@'; \
|
||||
set = JISX0201_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && outptr == outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = buf[0]; \
|
||||
} \
|
||||
else \
|
||||
{ \
|
||||
written = ucs4_to_jisx0208 (ch, buf, 2); \
|
||||
if (written != UNKNOWN_10646_CHAR) \
|
||||
{ \
|
||||
/* We use JIS X 0208. */ \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '$'; \
|
||||
*outptr++ = 'B'; \
|
||||
set = JISX0208_1983_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && outptr + 2 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = buf[0]; \
|
||||
*outptr++ = buf[1]; \
|
||||
} \
|
||||
else if (var == iso2022jp) \
|
||||
{ \
|
||||
/* We have no other choice. */ \
|
||||
result = GCONV_ILLEGAL_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
else \
|
||||
{ \
|
||||
written = ucs4_to_jisx0208 (ch, buf, 2); \
|
||||
if (written != UNKNOWN_10646_CHAR) \
|
||||
{ \
|
||||
/* We use JIS X 0212. */ \
|
||||
if (outptr + 4 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '$'; \
|
||||
*outptr++ = '('; \
|
||||
*outptr++ = 'D'; \
|
||||
set = JISX0212_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && outptr + 2 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = buf[0]; \
|
||||
*outptr++ = buf[1]; \
|
||||
} \
|
||||
else \
|
||||
{ \
|
||||
written = ucs4_to_gb2312 (ch, buf, 2); \
|
||||
if (written != UNKNOWN_10646_CHAR) \
|
||||
{ \
|
||||
/* We use GB 2312. */ \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '$'; \
|
||||
*outptr++ = 'A'; \
|
||||
set = GB2312_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST && outptr + 2 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = buf[0]; \
|
||||
*outptr++ = buf[1]; \
|
||||
} \
|
||||
else \
|
||||
{ \
|
||||
written = ucs4_to_ksc5601 (ch, buf, 2); \
|
||||
if (written != UNKNOWN_10646_CHAR) \
|
||||
{ \
|
||||
/* We use KSC 5601. */ \
|
||||
if (outptr + 4 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
*outptr++ = ESC; \
|
||||
*outptr++ = '$'; \
|
||||
*outptr++ = '('; \
|
||||
*outptr++ = 'C'; \
|
||||
set = KSC5601_set; \
|
||||
\
|
||||
if (NEED_LENGTH_TEST \
|
||||
&& outptr + 2 > outbufend) \
|
||||
{ \
|
||||
result = GCONV_FULL_OUTPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
*outptr++ = buf[0]; \
|
||||
*outptr++ = buf[1]; \
|
||||
} \
|
||||
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 set
|
||||
#include <iconv/loop.c>
|
||||
|
||||
|
||||
/* Now define the toplevel functions. */
|
||||
#include <iconv/skeleton.c>
|
||||
|
|
@ -152,10 +152,20 @@ gconv_init (struct gconv_step *step)
|
|||
new_data->var = var;
|
||||
step->data = new_data;
|
||||
|
||||
step->min_needed_from = MIN_NEEDED_FROM;
|
||||
step->max_needed_from = MIN_NEEDED_FROM;
|
||||
step->min_needed_to = MIN_NEEDED_TO;
|
||||
step->max_needed_to = MIN_NEEDED_TO;
|
||||
if (var == from_iso646)
|
||||
{
|
||||
step->min_needed_from = MIN_NEEDED_FROM;
|
||||
step->max_needed_from = MIN_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 = MIN_NEEDED_FROM;
|
||||
}
|
||||
|
||||
step->stateful = 0;
|
||||
|
||||
|
|
@ -408,7 +418,7 @@ gconv_end (struct gconv_step *data)
|
|||
#define LOOPFCT TO_LOOP
|
||||
#define BODY \
|
||||
{ \
|
||||
unsigned char ch = '\0'; \
|
||||
unsigned char ch; \
|
||||
int failure = GCONV_OK; \
|
||||
\
|
||||
ch = *((uint32_t *) inptr); \
|
||||
|
|
|
|||
|
|
@ -20,9 +20,9 @@
|
|||
|
||||
/* Get the conversion table. */
|
||||
#include <stdint.h>
|
||||
#include <iso8859-10.h>
|
||||
#define TABLES <iso8859-10.h>
|
||||
|
||||
#define CHARSET_NAME "ISO-8859-10//"
|
||||
#define HAS_HOLES 0 /* All 256 character are defined. */
|
||||
|
||||
#include <8bit-generic.c>
|
||||
#include <8bit-gap.c>
|
||||
|
|
|
|||
|
|
@ -20,9 +20,9 @@
|
|||
|
||||
/* Get the conversion table. */
|
||||
#include <stdint.h>
|
||||
#include <iso8859-5.h>
|
||||
#define TABLES <iso8859-5.h>
|
||||
|
||||
#define CHARSET_NAME "ISO-8859-5//"
|
||||
#define HAS_HOLES 0 /* All 256 character are defined. */
|
||||
|
||||
#include <8bit-generic.c>
|
||||
#include <8bit-gap.c>
|
||||
|
|
|
|||
|
|
@ -20,9 +20,9 @@
|
|||
|
||||
/* Get the conversion table. */
|
||||
#include <stdint.h>
|
||||
#include <iso8859-7.h>
|
||||
#define TABLES <iso8859-7.h>
|
||||
|
||||
#define CHARSET_NAME "ISO-8859-7//"
|
||||
#define HAS_HOLES 1 /* Not all 256 character are defined. */
|
||||
|
||||
#include <8bit-generic.c>
|
||||
#include <8bit-gap.c>
|
||||
|
|
|
|||
|
|
@ -20,9 +20,9 @@
|
|||
|
||||
/* Get the conversion table. */
|
||||
#include <stdint.h>
|
||||
#include <iso8859-8.h>
|
||||
#define TABLES <iso8859-8.h>
|
||||
|
||||
#define CHARSET_NAME "ISO-8859-8//"
|
||||
#define HAS_HOLES 1 /* Not all 256 character are defined. */
|
||||
|
||||
#include <8bit-generic.c>
|
||||
#include <8bit-gap.c>
|
||||
|
|
|
|||
3442
iconvdata/ksc5601.c
3442
iconvdata/ksc5601.c
File diff suppressed because it is too large
Load Diff
|
|
@ -27,46 +27,62 @@
|
|||
#include <gconv.h>
|
||||
#include <stdint.h>
|
||||
|
||||
/* Structure to map from UCS to KSC. This structure should be packed
|
||||
on all platforms. */
|
||||
struct map
|
||||
{
|
||||
uint16_t ucs;
|
||||
char val[2];
|
||||
};
|
||||
|
||||
/* Conversion table. */
|
||||
extern const uint16_t __ksc5601_hangul_to_ucs[KSC5601_HANGUL];
|
||||
extern const uint16_t __ksc5601_sym_to_ucs[];
|
||||
extern const uint16_t __ksc5601_sym_from_ucs[KSC5601_SYMBOL][2];
|
||||
extern const struct map __ksc5601_sym_from_ucs[KSC5601_SYMBOL];
|
||||
extern const uint16_t __ksc5601_hanja_to_ucs[KSC5601_HANJA];
|
||||
extern const uint16_t __ksc5601_hanja_from_ucs[KSC5601_HANJA][2];
|
||||
extern const struct map __ksc5601_hanja_from_ucs[KSC5601_HANJA];
|
||||
|
||||
|
||||
static inline uint32_t
|
||||
ksc5601_to_ucs4 (uint16_t s)
|
||||
ksc5601_to_ucs4 (const unsigned char **s, size_t avail, unsigned char offset)
|
||||
{
|
||||
unsigned char ch = s / 256;
|
||||
unsigned char ch = *(*s);
|
||||
unsigned char ch2;
|
||||
int idx;
|
||||
|
||||
/* row 94(0x7e) and row 41(0x49) are user-defined area in KS C 5601 */
|
||||
|
||||
if (ch <= 0x20 || ch >= 0x7e || ch == 0x49)
|
||||
if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) >= 0x7e
|
||||
|| (ch - offset) == 0x49)
|
||||
return UNKNOWN_10646_CHAR;
|
||||
|
||||
ch2 = s % 256;
|
||||
if (ch2 <= 0x20 || ch2 >= 0x7f)
|
||||
if (avail < 2)
|
||||
return 0;
|
||||
|
||||
ch2 = (*s)[1];
|
||||
if (ch2 < offset || (ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
|
||||
return UNKNOWN_10646_CHAR;
|
||||
|
||||
idx = (ch - 0x21) * 94 + (ch2 - 0x21);
|
||||
idx = (ch - offset - 0x21) * 94 + (ch2 - offset - 0x21);
|
||||
|
||||
/* 1410 = 15 * 94 , 3760 = 40 * 94
|
||||
Hangul in KS C 5601 : row 16 - row 40 */
|
||||
|
||||
(*s) += 2;
|
||||
|
||||
if (idx >= 1410 && idx < 3760)
|
||||
return __ksc5601_hangul_to_ucs[idx-1410];
|
||||
return (__ksc5601_hangul_to_ucs[idx - 1410]
|
||||
?: ((*s) -= 2, UNKNOWN_10646_CHAR));
|
||||
else if (idx > 3854)
|
||||
/* Hanja : row 42 - row 93 : 3854 = 94 * (42-1) */
|
||||
return __ksc5601_hanja_to_ucs[idx-3854];
|
||||
return (__ksc5601_hanja_to_ucs[idx - 3854]
|
||||
?: ((*s) -= 2, UNKNOWN_10646_CHAR));
|
||||
else
|
||||
return __ksc5601_sym_to_ucs[idx] ?: UNKNOWN_10646_CHAR;
|
||||
return __ksc5601_sym_to_ucs[idx] ?: ((*s) -= 2, UNKNOWN_10646_CHAR);
|
||||
}
|
||||
|
||||
static inline size_t
|
||||
ucs4_to_ksc5601_hangul (uint32_t wch, uint16_t *s)
|
||||
ucs4_to_ksc5601_hangul (uint32_t wch, unsigned char **s, size-t avail)
|
||||
{
|
||||
int l = 0;
|
||||
int m;
|
||||
|
|
@ -75,23 +91,30 @@ ucs4_to_ksc5601_hangul (uint32_t wch, uint16_t *s)
|
|||
|
||||
while (l <= u)
|
||||
{
|
||||
try = (uint32_t) __ksc5601_hangul_to_ucs[m=(l+u)/2];
|
||||
m = (l + u) / 2;
|
||||
try = (uint32_t) __ksc5601_hangul_to_ucs[m];
|
||||
if (try > wch)
|
||||
u = m - 1;
|
||||
else if (try < wch)
|
||||
l= m + 1;
|
||||
else
|
||||
{
|
||||
*s = (uint16_t) ((m / 94) * 256 + m % 94 + 0x3021) ;
|
||||
if (avail < 2)
|
||||
return 0;
|
||||
|
||||
*(*s)++ = (m / 94) + 0x30;
|
||||
*(*s)++ = (m % 94) + 0x21;
|
||||
|
||||
return 2;
|
||||
}
|
||||
}
|
||||
return 0;
|
||||
|
||||
return UNKNOWN_10646_CHAR;
|
||||
}
|
||||
|
||||
|
||||
static inline size_t
|
||||
ucs4_to_ksc5601_hanja (uint32_t wch, uint16_t *s)
|
||||
ucs4_to_ksc5601_hanja (uint32_t wch, unsigned char **s, size_t avail)
|
||||
{
|
||||
int l = 0;
|
||||
int m;
|
||||
|
|
@ -101,22 +124,28 @@ ucs4_to_ksc5601_hanja (uint32_t wch, uint16_t *s)
|
|||
while (l <= u)
|
||||
{
|
||||
m = (l + u) / 2;
|
||||
try = (uint32_t) __ksc5601_hanja_from_ucs[m][0];
|
||||
try = (uint32_t) __ksc5601_hanja_from_ucs[m].ucs;
|
||||
if (try > wch)
|
||||
u=m-1;
|
||||
else if (try < wch)
|
||||
l = m + 1;
|
||||
else
|
||||
{
|
||||
*s = __ksc5601_hanja_from_ucs[m][1];
|
||||
if (avail < 2)
|
||||
return 0;
|
||||
|
||||
*(*s)++ = __ksc5601_hanja_from_ucs[m].val[0];
|
||||
*(*s)++ = __ksc5601_hanja_from_ucs[m].val[1];
|
||||
|
||||
return 2;
|
||||
}
|
||||
}
|
||||
return 0;
|
||||
|
||||
return UNKNOWN_10646_CHAR;
|
||||
}
|
||||
|
||||
static inline size_t
|
||||
ucs4_to_ksc5601_sym (uint32_t wch, uint16_t *s)
|
||||
ucs4_to_ksc5601_sym (uint32_t wch, unsigned char **s, size_t avail)
|
||||
{
|
||||
int l = 0;
|
||||
int m;
|
||||
|
|
@ -126,32 +155,37 @@ ucs4_to_ksc5601_sym (uint32_t wch, uint16_t *s)
|
|||
while (l <= u)
|
||||
{
|
||||
m = (l + u) / 2;
|
||||
try = __ksc5601_sym_from_ucs[m][0];
|
||||
try = __ksc5601_sym_from_ucs[m].ucs;
|
||||
if (try > wch)
|
||||
u = m - 1;
|
||||
else if (try < wch)
|
||||
l = m + 1;
|
||||
else
|
||||
{
|
||||
*s = __ksc5601_sym_from_ucs[m][1];
|
||||
if (avail < 2)
|
||||
return 0;
|
||||
|
||||
*(*s)++ = __ksc5601_sym_from_ucs[m].val[0];
|
||||
*(*s)++ = __ksc5601_sym_from_ucs[m].val[1];
|
||||
|
||||
return 2;
|
||||
}
|
||||
}
|
||||
return 0;
|
||||
|
||||
return UNKNOWN_10646_CHAR;
|
||||
}
|
||||
|
||||
|
||||
static inline size_t
|
||||
ucs4_to_ksc5601 (uint32_t ch, uint16_t *s)
|
||||
ucs4_to_ksc5601 (uint32_t wch, unsigned char **s, size_t avail)
|
||||
{
|
||||
*s = (uint16_t) UNKNOWN_10646_CHAR; /* FIXIT */
|
||||
|
||||
if (ch >= 0xac00 && ch <= 0xd7a3)
|
||||
return ucs4_to_ksc5601_hangul (ch, s);
|
||||
else if (ch >= 0x4e00 && ch <= 0x9fff || ch >= 0xf900 && ch <= 0xfa0b)
|
||||
return ucs4_to_ksc5601_hanja (ch, s);
|
||||
if (wch >= 0xac00 && wch <= 0xd7a3)
|
||||
return ucs4_to_ksc5601_hangul (wch, (uint16_t *) s);
|
||||
else if ((wch >= 0x4e00 && wch <= 0x9fff)
|
||||
|| (wch >= 0xf900 && wch <= 0xfa0b))
|
||||
return ucs4_to_ksc5601_hanja (wch, s, avail);
|
||||
else
|
||||
return ucs4_to_ksc5601_sym (ch, s);
|
||||
return ucs4_to_ksc5601_sym (wch, s, avail);
|
||||
}
|
||||
|
||||
#endif /* ksc5601.h */
|
||||
|
|
|
|||
|
|
@ -2689,11 +2689,19 @@ uhc_from_ucs4 (uint32_t ch, unsigned char *cp)
|
|||
+ (ch < 0xa1 \
|
||||
? (ch - 0x81) * 178 \
|
||||
: 5696 + (ch - 0xa1) * 84)]; \
|
||||
\
|
||||
if (ch == 0) \
|
||||
{ \
|
||||
/* This is an illegal character. */ \
|
||||
result = GCONV_ILLEGAL_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
inptr += 2; \
|
||||
} \
|
||||
else \
|
||||
{ \
|
||||
ch = ksc5601_to_ucs4 ((ch * 256 + ch2) & 0x7f7f); \
|
||||
\
|
||||
ch = ksc5601_to_ucs4 (&inptr, 2, 0x80);
|
||||
if (ch == UNKNOWN_10646_CHAR) \
|
||||
{ \
|
||||
/* Illegal. */ \
|
||||
|
|
@ -2701,15 +2709,6 @@ uhc_from_ucs4 (uint32_t ch, unsigned char *cp)
|
|||
break; \
|
||||
} \
|
||||
} \
|
||||
\
|
||||
if (ch == 0) \
|
||||
{ \
|
||||
/* This is an illegal character. */ \
|
||||
result = GCONV_ILLEGAL_INPUT; \
|
||||
break; \
|
||||
} \
|
||||
\
|
||||
inptr += 2; \
|
||||
} \
|
||||
\
|
||||
*((uint32_t *) outptr)++ = ch; \
|
||||
|
|
|
|||
24
io/fcntl.h
24
io/fcntl.h
|
|
@ -61,8 +61,12 @@ extern int __open64 __P ((__const char *__file, int __oflag, ...));
|
|||
#ifndef __USE_FILE_OFFSET64
|
||||
extern int open __P ((__const char *__file, int __oflag, ...));
|
||||
#else
|
||||
extern int open __P ((__const char *__file, int __oflag, ...))
|
||||
__asm__ ("open64");
|
||||
# ifdef __REDIRECT
|
||||
extern int __REDIRECT (open, __P ((__const char *__file, int __oflag, ...)),
|
||||
open64);
|
||||
# else
|
||||
# define open open64
|
||||
# endif
|
||||
#endif
|
||||
#ifdef __USE_LARGEFILE64
|
||||
extern int open64 __P ((__const char *__file, int __oflag, ...));
|
||||
|
|
@ -74,8 +78,12 @@ extern int open64 __P ((__const char *__file, int __oflag, ...));
|
|||
#ifndef __USE_FILE_OFFSET64
|
||||
extern int creat __P ((__const char *__file, __mode_t __mode));
|
||||
#else
|
||||
extern int creat __P ((__const char *__file, __mode_t __mode))
|
||||
__asm__ ("creat64");
|
||||
# ifdef __REDIRECT
|
||||
extern int __REDIRECT (creat, __P ((__const char *__file, __mode_t __mode)),
|
||||
creat64);
|
||||
# else
|
||||
# define creat creat64
|
||||
# endif
|
||||
#endif
|
||||
#ifdef __USE_LARGEFILE64
|
||||
extern int creat64 __P ((__const char *__file, __mode_t __mode));
|
||||
|
|
@ -99,8 +107,12 @@ extern int creat64 __P ((__const char *__file, __mode_t __mode));
|
|||
# ifndef __USE_FILE_OFFSET64
|
||||
extern int lockf __P ((int __fd, int __cmd, __off_t __len));
|
||||
# else
|
||||
extern int lockf __P ((int __fd, int __cmd, __off64_t __len))
|
||||
__asm__ ("lockf64");
|
||||
# ifdef __REDIRECT
|
||||
extern int __REDIRECT (lockf, __P ((int __fd, int __cmd, __off64_t __len)),
|
||||
lockf64);
|
||||
# else
|
||||
# define lockf lockf64
|
||||
# endif
|
||||
# endif
|
||||
# ifdef __USE_LARGEFILE64
|
||||
extern int lockf64 __P ((int __fd, int __cmd, __off64_t __len));
|
||||
|
|
|
|||
Loading…
Reference in New Issue