The 'mbrtowc' module emulation, based on the system's mbtowc(), is more
interesting:
- sizeof(mbstate_t) has to be >= 4 for UTF-8, but on IRIX 6.5 mbstate_t is
defined to 'char'. (Thanks to Paul for idea to verify this assumption :-)).
So we have to override this type.
- mbtowc() does not report to the caller a distinction between incomplete and
invalid multibyte sequence. But mbrtowc() has to do so. How to do so, is
tricky.
- What should the return value of mbrtowc(&wc, s, 0) be? According to the
spec, I would think (size_t)(-2). But glibc returns 0. Hmm...
- glibc's mbtowc() implementation for UTF-8 is buggy.
2008-12-18 Bruno Haible <br...@clisp.org>
New module 'mbrtowc'.
* lib/wchar.in.h (mbstate_t): Override when the system does not have
mbsinit and mbrtowc.
(mbrtowc): New declaration.
* lib/mbrtowc.c: New file.
* m4/mbrtowc.m4 (gl_FUNC_MBRTOWC, gl_PREREQ_MBRTOWC): New macros.
* modules/mbrtowc: New file.
* m4/wchar.m4 (gl_WCHAR_H_DEFAULTS): Initialize GNULIB_MBRTOWC and
HAVE_MBRTOWC.
* modules/wchar (Makefile.am): Substitute GNULIB_MBRTOWC and
HAVE_MBRTOWC.
* doc/posix-functions/mbrtowc.texi: Document the new module.
================================ lib/mbrtowc.c ================================
/* Convert multibyte character to wide character.
Copyright (C) 1999-2002, 2005-2008 Free Software Foundation, Inc.
Written by Bruno Haible <br...@clisp.org>, 2008.
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 3 of the License, or
(at your option) any later version.
This program 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 General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>. */
#include <config.h>
/* Specification. */
#include <wchar.h>
#include <errno.h>
#include <stdlib.h>
#include "localcharset.h"
#include "streq.h"
static char internal_state[4];
size_t
mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps)
{
char *pstate = (char *)ps;
if (pstate == NULL)
pstate = internal_state;
if (s == NULL)
{
pwc = NULL;
s = "";
n = 1;
}
if (n == 0)
return (size_t)(-2);
/* Here n > 0. */
{
size_t nstate = pstate[0];
char buf[4];
const char *p;
size_t m;
switch (nstate)
{
case 0:
p = s;
m = n;
break;
case 3:
buf[2] = pstate[3];
/*FALLTHROUGH*/
case 2:
buf[1] = pstate[2];
/*FALLTHROUGH*/
case 1:
buf[0] = pstate[1];
p = buf;
m = nstate;
buf[m++] = s[0];
if (n >= 2 && m < 4)
{
buf[m++] = s[1];
if (n >= 3 && m < 4)
buf[m++] = s[2];
}
break;
default:
errno = EINVAL;
return (size_t)(-1);
}
/* Here 0 < m ≤ 4. */
#if __GLIBC__
/* Work around bug <http://sourceware.org/bugzilla/show_bug.cgi?id=9674> */
mbtowc (NULL, NULL, 0);
#endif
{
int res = mbtowc (pwc, p, m);
if (res >= 0)
{
if (pwc != NULL && ((*pwc == 0) != (res == 0)))
abort ();
if (nstate >= (res > 0 ? res : 1))
abort ();
res -= nstate;
pstate[0] = 0;
return res;
}
/* mbtowc does not distinguish between invalid and incomplete multibyte
sequences. But mbrtowc needs to make this distinction.
There are two possible approaches:
- Use iconv() and its return value.
- Use built-in knowledge about the possible encodings.
Given the low quality of implementation of iconv() on the systems that
lack mbrtowc(), we use the second approach.
The possible encodings are:
- 8-bit encodings,
- EUC-JP, EUC-KR, GB2312, EUC-TW, BIG5, SJIS,
- UTF-8.
Use specialized code for each. */
if (m >= 4 || m >= MB_CUR_MAX)
goto invalid;
/* Here MB_CUR_MAX > 1 and 0 < m < 4. */
{
const char *encoding = locale_charset ();
if (STREQ (encoding, "UTF-8", 'U', 'T', 'F', '-', '8', 0, 0, 0, 0))
{
/* Cf. unistr/u8-mblen.c. */
unsigned char c = (unsigned char) p[0];
if (c >= 0xc2)
{
if (c < 0xe0)
{
if (m == 1)
goto incomplete;
}
else if (c < 0xf0)
{
if (m == 1)
goto incomplete;
if (m == 2)
{
unsigned char c2 = (unsigned char) p[1];
if ((c2 ^ 0x80) < 0x40
&& (c >= 0xe1 || c2 >= 0xa0)
&& (c != 0xed || c2 < 0xa0))
goto incomplete;
}
}
else if (c <= 0xf4)
{
if (m == 1)
goto incomplete;
else /* m == 2 || m == 3 */
{
unsigned char c2 = (unsigned char) p[1];
if ((c2 ^ 0x80) < 0x40
&& (c >= 0xf1 || c2 >= 0x90)
&& (c < 0xf4 || (c == 0xf4 && c2 < 0x90)))
{
if (m == 2)
goto incomplete;
else /* m == 3 */
{
unsigned char c3 = (unsigned char) p[2];
if ((c3 ^ 0x80) < 0x40)
goto incomplete;
}
}
}
}
}
goto invalid;
}
/* As a reference for this code, you can use the GNU libiconv
implementation. Look for uses of the RET_TOOFEW macro. */
if (STREQ (encoding, "EUC-JP", 'E', 'U', 'C', '-', 'J', 'P', 0, 0, 0))
{
if (m == 1)
{
unsigned char c = (unsigned char) p[0];
if ((c >= 0xa1 && c < 0xff) || c == 0x8e || c == 0x8f)
goto incomplete;
}
if (m == 2)
{
unsigned char c = (unsigned char) p[0];
if (c == 0x8f)
{
unsigned char c2 = (unsigned char) p[1];
if (c2 >= 0xa1 && c2 < 0xff)
goto incomplete;
}
}
goto invalid;
}
if (STREQ (encoding, "EUC-KR", 'E', 'U', 'C', '-', 'K', 'R', 0, 0, 0)
|| STREQ (encoding, "GB2312", 'G', 'B', '2', '3', '1', '2', 0, 0, 0)
|| STREQ (encoding, "BIG5", 'B', 'I', 'G', '5', 0, 0, 0, 0, 0))
{
if (m == 1)
{
unsigned char c = (unsigned char) p[0];
if (c >= 0xa1 && c < 0xff)
goto incomplete;
}
goto invalid;
}
if (STREQ (encoding, "EUC-TW", 'E', 'U', 'C', '-', 'T', 'W', 0, 0, 0))
{
if (m == 1)
{
unsigned char c = (unsigned char) p[0];
if ((c >= 0xa1 && c < 0xff) || c == 0x8e)
goto incomplete;
}
else /* m == 2 || m == 3 */
{
unsigned char c = (unsigned char) p[0];
if (c == 0x8e)
goto incomplete;
}
goto invalid;
}
if (STREQ (encoding, "SJIS", 'S', 'J', 'I', 'S', 0, 0, 0, 0, 0))
{
if (m == 1)
{
unsigned char c = (unsigned char) p[0];
if ((c >= 0x81 && c <= 0x9f) || (c >= 0xe0 && c <= 0xea)
|| (c >= 0xf0 && c <= 0xf9))
goto incomplete;
}
goto invalid;
}
/* An unknown multibyte encoding. */
goto incomplete;
}
incomplete:
{
size_t k = nstate;
/* Here 0 < k < m < 4. */
pstate[++k] = s[0];
if (k < m)
pstate[++k] = s[1];
if (k != m)
abort ();
}
pstate[0] = m;
return (size_t)(-2);
invalid:
errno = EILSEQ;
/* The conversion state is undefined, says POSIX. */
return (size_t)(-1);
}
}
}
=============================== modules/mbrtowc ===============================
Description:
mbrtowc() function: convert multibyte character to wide character.
Files:
lib/mbrtowc.c
m4/mbrtowc.m4
m4/mbstate_t.m4
Depends-on:
wchar
localcharset
streq
configure.ac:
gl_FUNC_MBRTOWC
gl_WCHAR_MODULE_INDICATOR([mbrtowc])
Makefile.am:
Include:
<wchar.h>
License:
LGPL
Maintainer:
Bruno Haible
===============================================================================
--- lib/wchar.in.h.orig 2008-12-19 03:15:03.000000000 +0100
+++ lib/wchar.in.h 2008-12-19 03:13:37.000000000 +0100
@@ -71,6 +71,16 @@
#endif
+/* Override mbstate_t if it is too small.
+ On IRIX 6.5, sizeof (mbstate_t) == 1, which is not sufficient for
+ implementing mbrtowc for encodings like UTF-8. */
+#if !(@HAVE_MBSINIT@ && @HAVE_MBRTOWC@)
+typedef int rpl_mbstate_t;
+# undef mbstate_t
+# define mbstate_t rpl_mbstate_t
+#endif
+
+
/* Convert a single-byte character to a wide character. */
#if @GNULIB_BTOWC@
# if !...@have_btowc@
@@ -113,6 +123,20 @@
#endif
+/* Convert a multibyte character to a wide character. */
+#if @GNULIB_MBRTOWC@
+# if !...@have_mbrtowc@
+extern size_t mbrtowc (wchar_t *pwc, const char *s, size_t n, mbstate_t *ps);
+# endif
+#elif defined GNULIB_POSIXCHECK
+# undef mbrtowc
+# define mbrtowc(w,s,n,p) \
+ (GL_LINK_WARNING ("mbrtowc is unportable - " \
+ "use gnulib module mbrtowc for portability"), \
+ mbrtowc (w, s, n, p))
+#endif
+
+
/* Return the number of screen columns needed for WC. */
#if @GNULIB_WCWIDTH@
# if @REPLACE_WCWIDTH@
--- m4/mbrtowc.m4.orig 2008-12-19 03:15:03.000000000 +0100
+++ m4/mbrtowc.m4 2008-12-19 03:13:37.000000000 +0100
@@ -1,9 +1,29 @@
-# mbrtowc.m4 serial 9
+# mbrtowc.m4 serial 10
dnl Copyright (C) 2001-2002, 2004-2005, 2008 Free Software Foundation, Inc.
dnl This file is free software; the Free Software Foundation
dnl gives unlimited permission to copy and/or distribute it,
dnl with or without modifications, as long as this notice is preserved.
+AC_DEFUN([gl_FUNC_MBRTOWC],
+[
+ AC_REQUIRE([gl_WCHAR_H_DEFAULTS])
+
+ AC_REQUIRE([AC_TYPE_MBSTATE_T])
+ AC_CHECK_FUNCS_ONCE([mbrtowc])
+ if test $ac_cv_func_mbrtowc = no; then
+ HAVE_MBRTOWC=0
+ gl_REPLACE_WCHAR_H
+ AC_LIBOBJ([mbrtowc])
+ gl_PREREQ_MBRTOWC
+ fi
+])
+
+# Prerequisites of lib/mbrtowc.c.
+AC_DEFUN([gl_PREREQ_MBRTOWC], [
+ :
+])
+
+
dnl From Paul Eggert
dnl This override of an autoconf macro can be removed when autoconf 2.60 or
--- m4/wchar.m4.orig 2008-12-19 03:15:03.000000000 +0100
+++ m4/wchar.m4 2008-12-19 03:13:37.000000000 +0100
@@ -64,11 +64,13 @@
GNULIB_BTOWC=0; AC_SUBST([GNULIB_BTOWC])
GNULIB_WCTOB=0; AC_SUBST([GNULIB_WCTOB])
GNULIB_MBSINIT=0; AC_SUBST([GNULIB_MBSINIT])
+ GNULIB_MBRTOWC=0; AC_SUBST([GNULIB_MBRTOWC])
GNULIB_WCWIDTH=0; AC_SUBST([GNULIB_WCWIDTH])
dnl Assume proper GNU behavior unless another module says otherwise.
HAVE_BTOWC=1; AC_SUBST([HAVE_BTOWC])
HAVE_WCTOB=1; AC_SUBST([HAVE_WCTOB])
HAVE_MBSINIT=1; AC_SUBST([HAVE_MBSINIT])
+ HAVE_MBRTOWC=1; AC_SUBST([HAVE_MBRTOWC])
HAVE_DECL_WCWIDTH=1; AC_SUBST([HAVE_DECL_WCWIDTH])
REPLACE_WCWIDTH=0; AC_SUBST([REPLACE_WCWIDTH])
WCHAR_H=''; AC_SUBST([WCHAR_H])
--- modules/wchar.orig 2008-12-19 03:15:03.000000000 +0100
+++ modules/wchar 2008-12-19 03:13:37.000000000 +0100
@@ -28,11 +28,13 @@
-e 's|@''GNULIB_BTOWC''@|$(GNULIB_BTOWC)|g' \
-e 's|@''GNULIB_WCTOB''@|$(GNULIB_WCTOB)|g' \
-e 's|@''GNULIB_MBSINIT''@|$(GNULIB_MBSINIT)|g' \
+ -e 's|@''GNULIB_MBRTOWC''@|$(GNULIB_MBRTOWC)|g' \
-e 's|@''GNULIB_WCWIDTH''@|$(GNULIB_WCWIDTH)|g' \
-e 's|@''HAVE_WINT_T''@|$(HAVE_WINT_T)|g' \
-e 's|@''HAVE_BTOWC''@|$(HAVE_BTOWC)|g' \
-e 's|@''HAVE_WCTOB''@|$(HAVE_WCTOB)|g' \
-e 's|@''HAVE_MBSINIT''@|$(HAVE_MBSINIT)|g' \
+ -e 's|@''HAVE_MBRTOWC''@|$(HAVE_MBRTOWC)|g' \
-e 's|@''HAVE_DECL_WCWIDTH''@|$(HAVE_DECL_WCWIDTH)|g' \
-e 's|@''REPLACE_WCWIDTH''@|$(REPLACE_WCWIDTH)|g' \
-e '/definition of GL_LINK_WARNING/r $(LINK_WARNING_H)' \
--- doc/posix-functions/mbrtowc.texi.orig 2008-12-19 03:15:03.000000000
+0100
+++ doc/posix-functions/mbrtowc.texi 2008-12-19 03:13:37.000000000 +0100
@@ -4,18 +4,18 @@
POSIX specification:
@url{http://www.opengroup.org/onlinepubs/9699919799/functions/mbrtowc.html}
-Gnulib module: ---
+Gnulib module: mbrtowc
Portability problems fixed by Gnulib:
@itemize
+...@item
+This function is missing on some platforms:
+HP-UX 11, IRIX 6.5, Solaris 2.6, mingw, Interix 3.5.
@end itemize
Portability problems not fixed by Gnulib:
@itemize
@item
-This function is missing on some platforms:
-HP-UX 11, IRIX 6.5, Solaris 2.6, mingw, Interix 3.5.
-...@item
On Windows platforms, @code{wchar_t} is a 16-bit type and therefore cannot
accommodate all Unicode characters.
@end itemize
