1998-04-13 09:11  Ulrich Drepper  <drepper@cygnus.com>

	* iconvdata/eucjp.c: New file.
	* iconvdata/euccn.c: New file.
	* iconvdata/gb2312.h: New file.
	* iconvdata/gb2312.c: New file.
	* iconvdata/Makefile: Add rules for EUC-JP and EUC-CN.
	* iconvdata/gconv-modules: Likewise.

	* iconvdata/euckr.c: Use stdint.h instead of inttypes.h.
This commit is contained in:
Ulrich Drepper 1998-04-13 09:17:55 +00:00
parent 8acdf4fd69
commit 40b4c81d4e
7 changed files with 5083 additions and 4 deletions

View File

@ -1,3 +1,14 @@
1998-04-13 09:11 Ulrich Drepper <drepper@cygnus.com>
* iconvdata/eucjp.c: New file.
* iconvdata/euccn.c: New file.
* iconvdata/gb2312.h: New file.
* iconvdata/gb2312.c: New file.
* iconvdata/Makefile: Add rules for EUC-JP and EUC-CN.
* iconvdata/gconv-modules: Likewise.
* iconvdata/euckr.c: Use stdint.h instead of inttypes.h.
1998-04-12 Ulrich Drepper <drepper@cygnus.com>
* sysdeps/unix/sysv/linux/powerpc/dl-sysdep.c

View File

@ -26,7 +26,8 @@ modules := ISO8859-1 ISO8859-2 ISO8859-3 ISO8859-4 ISO8859-5 \
ISO8859-6 ISO8859-7 ISO8859-8 ISO8859-9 ISO8859-10 \
T.61 ISO_6937 SJIS KOI-8 KOI8-R LATIN-GREEK LATIN-GREEK-1 \
HP-ROMAN8 EBCDIC-AT-DE EBCDIC-AT-DE-A EBCDIC-CA-FR \
EUC-KR UHC JOHAB libJIS libKSC ISO646 BIG5
EUC-KR UHC JOHAB libJIS libKSC ISO646 BIG5 EUC-JP libGB \
EUC-CN
modules.so := $(addsuffix .so, $(modules))
@ -58,8 +59,11 @@ EUC-KR-routines := euckr
JOHAB-routines := johab
UHC-routines := uhc
BIG5-routines := big5
EUC-JP-routines := eucjp
EUC-CN-routines := euccn
libJIS-routines := jis0201 jis0208 jis0212
libKSC-routines := ksc5601
libGB-routines := gb2312
LDFLAGS-EUC-KR.so = -Wl,-rpath,$(gconvdir)
$(objpfx)EUC-KR.so: $(objpfx)libKSC.so
@ -67,9 +71,14 @@ LDFLAGS-JOHAB.so = -Wl,-rpath,$(gconvdir)
$(objpfx)JOHAB.so: $(objpfx)libKSC.so
LDFLAGS-UHC.so = -Wl,-rpath,$(gconvdir)
$(objpfx)UHC.so: $(objpfx)libKSC.so
LDFLAGS-EUC-JP.so = -Wl,-rpath,$(gconvdir)
$(objpfx)EUC-JP.so: $(objpfx)libJIS.so
LDFLAGS-EUC-CN.so = -Wl,-rpath,$(gconvdir)
$(objpfx)EUC-CN.so: $(objpfx)libGB.so
LDFLAGS-libJIS.so = -Wl,-soname,$(@F)
LDFLAGS-libKSC.so = -Wl,-soname,$(@F)
LDFLAGS-libGB.so = -Wl,-soname,$(@F)
distribute := 8bit-generic.c 8bit-gap.c gap.pl gaptab.pl gconv-modules \
iso8859-1.c iso8859-2.c iso8859-3.c iso8859-4.c iso8859-5.c \
@ -79,7 +88,7 @@ distribute := 8bit-generic.c 8bit-gap.c gap.pl gaptab.pl gconv-modules \
latin-greek.h latin-greek-1.c latin-greek-1.h ebcdic-at-de.c \
ebcdic-at-de-a.c ebcdic-ca-fr.c jis0201.c jis0208.c jis0212.c \
extra-module.mk euckr.c johab.c uhc.c ksc5601.c ksc5601.h \
iso646.c big5.c
iso646.c big5.c eucjp.c gb2312.c gb2312.h euccn.c
# We build the transformation modules only when we build shared libs.
ifeq (yes,$(build-shared))

304
iconvdata/euccn.c Normal file
View File

@ -0,0 +1,304 @@
/* Mapping tables for EUC-CN handling.
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 <stdlib.h>
#include <string.h>
#include <wchar.h>
#include <gb2312.h>
/* Direction of the transformation. */
enum direction
{
illegal,
to_euccn,
from_euccn
};
struct euccn_data
{
enum direction dir;
};
int
gconv_init (struct gconv_step *step)
{
/* Determine which direction. */
struct euccn_data *new_data;
enum direction dir;
int result;
if (strcasestr (step->from_name, "EUC-CN") != NULL)
dir = from_euccn;
else if (strcasestr (step->to_name, "EUC-CN") != NULL)
dir = to_euccn;
else
dir = illegal;
result = GCONV_NOCONV;
if (dir != illegal
&& ((new_data
= (struct euccn_data *) malloc (sizeof (struct euccn_data)))
!= NULL))
{
new_data->dir = dir;
step->data = new_data;
result = GCONV_OK;
}
return result;
}
void
gconv_end (struct gconv_step *data)
{
free (data->data);
}
int
gconv (struct gconv_step *step, struct gconv_step_data *data,
const char *inbuf, size_t *inbufsize, size_t *written, int do_flush)
{
struct gconv_step *next_step = step + 1;
struct gconv_step_data *next_data = data + 1;
gconv_fct fct = next_step->fct;
size_t do_write;
int result;
/* 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 (do_flush)
{
do_write = 0;
/* Call the steps down the chain if there are any. */
if (data->is_last)
result = GCONV_OK;
else
{
struct gconv_step *next_step = step + 1;
struct gconv_step_data *next_data = data + 1;
result = (*fct) (next_step, next_data, NULL, 0, written, 1);
/* Clear output buffer. */
data->outbufavail = 0;
}
}
else
{
enum direction dir = ((struct euccn_data *) step->data)->dir;
do_write = 0;
do
{
result = GCONV_OK;
if (dir == from_euccn)
{
size_t inchars = *inbufsize;
size_t outwchars = data->outbufavail;
char *outbuf = data->outbuf;
size_t cnt = 0;
while (cnt < inchars
&& (outwchars + sizeof (wchar_t) <= data->outbufsize))
{
int inchar = (unsigned char) inbuf[cnt];
wchar_t ch;
if (inchar <= 0x7f)
ch = (wchar_t) inchar;
else if ((inchar <= 0xa0 || inchar > 0xfe)
&& inchar != 0x8e && inchar != 0x8f)
/* This is illegal. */
ch = L'\0';
else
{
/* Two or more byte character. First test whether the
next character is also available. */
const char *endp;
int inchar2;
if (cnt + 1 >= inchars)
{
/* The second character is not available. Store
the intermediate result. */
result = GCONV_INCOMPLETE_INPUT;
break;
}
inchar2 = (unsigned char) inbuf[++cnt];
/* All second bytes of a multibyte character must be
>= 0xa1. */
if (inchar2 < 0xa1)
{
/* This is an illegal character. */
--cnt;
result = GCONV_ILLEGAL_INPUT;
break;
}
/* This is code set 1: GB 2312-80. */
endp = &inbuf[cnt - 1];
ch = gb2312_to_ucs4 (&endp, 2, 0x80);
if (ch != L'\0')
++cnt;
if (ch == UNKNOWN_10646_CHAR)
ch = L'\0';
if (ch == L'\0')
--cnt;
}
if (ch == L'\0' && inbuf[cnt] != '\0')
{
/* This is an illegal character. */
result = GCONV_ILLEGAL_INPUT;
break;
}
*((wchar_t *) (outbuf + outwchars)) = ch;
++do_write;
outwchars += sizeof (wchar_t);
++cnt;
}
*inbufsize -= cnt;
data->outbufavail = outwchars;
}
else
{
size_t inwchars = *inbufsize;
size_t outchars = data->outbufavail;
char *outbuf = data->outbuf;
size_t cnt = 0;
int extra = 0;
while (inwchars >= cnt + sizeof (wchar_t)
&& outchars < data->outbufsize)
{
wchar_t ch = *((wchar_t *) (inbuf + cnt));
if (ch <= L'\x7f')
/* It's plain ASCII. */
outbuf[outchars] = ch;
else
{
/* Try the JIS character sets. */
size_t found;
found = ucs4_to_gb2312 (ch, &outbuf[outchars],
(data->outbufsize
- outchars));
if (found > 0)
{
/* It's a GB 2312 character, adjust it for
EUC-CN. */
outbuf[outchars++] += 0x80;
outbuf[outchars] += 0x80;
}
else if (found == 0)
{
/* We ran out of space. */
extra = 2;
break;
}
else
/* Illegal character. */
break;
}
++do_write;
++outchars;
cnt += sizeof (wchar_t);
}
*inbufsize -= cnt;
data->outbufavail = outchars;
if (outchars + extra < data->outbufsize)
{
/* If there is still room in the output buffer something
is wrong with the input. */
if (inwchars >= cnt + sizeof (wchar_t))
{
/* An error occurred. */
result = GCONV_ILLEGAL_INPUT;
break;
}
if (inwchars != cnt)
{
/* There are some unprocessed bytes at the end of the
input buffer. */
result = GCONV_INCOMPLETE_INPUT;
break;
}
}
}
if (result != GCONV_OK)
break;
if (data->is_last)
{
/* This is the last step. */
result = (*inbufsize > (dir == from_euccn
? 0 : sizeof (wchar_t) - 1)
? GCONV_FULL_OUTPUT : GCONV_EMPTY_INPUT);
break;
}
/* Status so far. */
result = GCONV_EMPTY_INPUT;
if (data->outbufavail > 0)
{
/* Call the functions below in the chain. */
size_t newavail = data->outbufavail;
result = (*fct) (next_step, next_data, data->outbuf, &newavail,
written, 0);
/* Correct the output buffer. */
if (newavail != data->outbufavail && newavail > 0)
{
memmove (data->outbuf,
&data->outbuf[data->outbufavail - newavail],
newavail);
data->outbufavail = newavail;
}
}
}
while (*inbufsize > 0 && result == GCONV_EMPTY_INPUT);
}
if (written != NULL && data->is_last)
*written = do_write;
return result;
}

View File

@ -19,7 +19,7 @@
Boston, MA 02111-1307, USA. */
#include <gconv.h>
#include <inttypes.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
#include <wchar.h>
@ -168,7 +168,7 @@ gconv (struct gconv_step *step, struct gconv_step_data *data,
/* 0xfe(->0x7e : row 94) and 0xc9(->0x59 : row 41) are user-defined areas */
else if ( inchar <= 0xa0 || inchar >= 0xfe || inchar == 0xc9)
else if ( inchar <= 0xa0 || inchar > 0xfe || inchar == 0xc9)
/* This is illegal. */
ch = L'\0';
else

4560
iconvdata/gb2312.c Normal file

File diff suppressed because it is too large Load Diff

185
iconvdata/gb2312.h Normal file
View File

@ -0,0 +1,185 @@
/* Access functions for GB2312 conversion.
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. */
#ifndef _GB2312_H
#define _GB2312_H 1
#include <gconv.h>
#include <stdint.h>
/* Conversion table. */
extern const uint16_t gb2312_to_ucs[];
static inline wchar_t
gb2312_to_ucs4 (const char **s, size_t avail, unsigned char offset)
{
unsigned char ch = *(*s);
unsigned char ch2;
int idx;
if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) > 0x77)
return UNKNOWN_10646_CHAR;
if (avail < 2)
return 0;
ch2 = (*s)[1];
if ((ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
return UNKNOWN_10646_CHAR;
idx = (ch - 0x21 - offset) * 94 + (ch2 - 0x21 - offset);
if (idx >= 0x1ff1)
return UNKNOWN_10646_CHAR;
(*s) += 2;
return gb2312_to_ucs[idx] ?: ((*s) -= 2, UNKNOWN_10646_CHAR);
}
extern const char gb2312_from_ucs4_tab1[][2];
extern const char gb2312_from_ucs4_tab2[][2];
extern const char gb2312_from_ucs4_tab3[][2];
extern const char gb2312_from_ucs4_tab4[][2];
extern const char gb2312_from_ucs4_tab5[][2];
extern const char gb2312_from_ucs4_tab6[][2];
extern const char gb2312_from_ucs4_tab7[][2];
extern const char gb2312_from_ucs4_tab8[][2];
extern const char gb2312_from_ucs4_tab9[][2];
static inline size_t
ucs4_to_gb2312 (wchar_t wch, char *s, size_t avail)
{
unsigned int ch = (unsigned int) wch;
char buf[2];
const char *cp = NULL;
if (ch < 0xa4)
cp = NULL;
else if (ch < 0x101)
cp = gb2312_from_ucs4_tab1[ch - 0xa4];
else if (ch == 0x113)
cp = "\x28\x25";
else if (ch == 0x11b)
cp = "\x28\x27";
else if (ch == 0x12b)
cp = "\x28\x29";
else if (ch == 0x14d)
cp = "\x28\x2d";
else if (ch == 0x16b)
cp = "\x28\x31";
else if (ch == 0x1ce)
cp = "\x28\x23";
else if (ch == 0x1d0)
cp = "\x28\x2b";
else if (ch == 0x1d2)
cp = "\x28\x2f";
else if (ch == 0x1d4)
cp = "\x28\x33";
else if (ch == 0x1d6)
cp = "\x28\x35";
else if (ch == 0x1d8)
cp = "\x28\x36";
else if (ch == 0x1da)
cp = "\x28\x37";
else if (ch == 0x1dc)
cp = "\x28\x38";
else if (ch == 0x2c7)
cp = "\x21\x26";
else if (ch == 0x2c9)
cp = "\x21\x25";
else if (ch >= 0x391 && ch <= 0x3c9)
cp = gb2312_from_ucs4_tab2[ch - 0x391];
else if (ch >= 0x401 && ch <= 0x451)
cp = gb2312_from_ucs4_tab3[ch - 0x401];
else if (ch >= 0x2015 && ch <= 0x203b)
cp = gb2312_from_ucs4_tab4[ch - 0x2015];
else if (ch >= 0x2103 && ch <= 0x22a5)
cp = gb2312_from_ucs4_tab5[ch - 0x2103];
else if (ch == 0x2313)
cp = "\x21\x50";
else if (ch >= 0x2460 && ch <= 0x249b)
cp = gb2312_from_ucs4_tab6[ch - 0x2460];
else if (ch >= 0x2500 && ch <= 0x254b)
{
buf[0] = '\x29';
buf[1] = '\x24' + (ch & 256);
cp = buf;
}
else if (ch == 0x25a0)
cp = "\x21\x76";
else if (ch == 0x25a1)
cp = "\x21\x75";
else if (ch == 0x25b2)
cp = "\x21\x78";
else if (ch == 0x25b3)
cp = "\x21\x77";
else if (ch == 0x25c6)
cp = "\x21\x74";
else if (ch == 0x25c7)
cp = "\x21\x73";
else if (ch == 0x25cb)
cp = "\x21\x70";
else if (ch == 0x25ce)
cp = "\x21\x72";
else if (ch == 0x25cf)
cp = "\x21\x71";
else if (ch == 0x2605)
cp = "\x21\x6f";
else if (ch == 0x2606)
cp = "\x21\x6e";
else if (ch == 0x2640)
cp = "\x21\x62";
else if (ch == 0x2642)
cp = "\x21\x61";
else if (ch >= 0x3000 && ch <= 0x3129)
cp = gb2312_from_ucs4_tab7[ch - 0x3000];
else if (ch >= 0x3220 && ch <= 0x3229)
{
buf[0] = '\x22';
buf[1] = '\x65' + (ch - 0x3220);
cp = buf;
}
else if (ch >= 0x4e00 && ch <= 0x9fa0)
cp = gb2312_from_ucs4_tab8[ch - 0x4e00];
else if (ch >= 0xff01 && ch <= 0xff5e)
cp = gb2312_from_ucs4_tab9[ch - 0xff01];
else if (ch == 0xffe0)
cp = "\x21\x69";
else if (ch == 0xffe1)
cp = "\x21\x6a";
else if (ch == 0xffe3)
cp = "\x23\x7e";
else if (ch == 0xffe5)
cp = "\x23\x24";
else
return UNKNOWN_10646_CHAR;
if (cp[1] != '\0' && avail < 2)
return 0;
s[0] = cp[0];
s[1] = cp[1];
return 2;
}
#endif /* gb2312.h */

View File

@ -226,3 +226,13 @@ alias BIGFIVE// BIG5//
alias BIG-5// BIG5//
module BIG5// ISO-10646/UCS4/ BIG5 1
module ISO-10646/UCS4/ BIG5// BIG5 1
# from to module cost
alias EUCJP// EUC-JP//
module EUC-JP// ISO-10646/UCS4/ EUC-JP 1
module ISO-10646/UCS4/ EUC-JP// EUC-JP 1
# from to module cost
alias EUCCN// EUC-CN//
module EUC-CN// ISO-10646/UCS4/ EUC-CN 1
module ISO-10646/UCS4/ EUC-CN// EUC-CN 1